Skip to main content

Full text of "Electric Welding and Welding Appliances"

See other formats


This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 

to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 

to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 

are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other maiginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 

publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing tliis resource, we liave taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 
We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain fivm automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attributionTht GoogXt "watermark" you see on each file is essential for in forming people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liabili^ can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http: //books .google .com/I 


of tbe 

mniverditi? of Midconein 






What Industry Owes to Chemical 

Science. By Richard B. Pilchbr and 
Frank Butler-Jonks, B. A. With an Intro- 
duction by Sir George Bbilby» LL.D., 
F.R.S. Crown 8vo. 

The Production and Treatment of 

Vegetable Oils. By T. W. Chalmers, 
B.Sc., A.M.I.Mech.E. Imperial 8vo. 9 
Folding Plates, 95 Illustrations. 

Paper Making and its Machinery. 

By T. W. Chalmers. Imperial 8vo. 

Electric Welding and Welding 

Appliances. By II. Carpmael. Imperial 
8vo. Illustrated. 





(On the Editorial Staff of "The Engineer") 





Printed in Great Britain. 

.? 3 ? 


HPT — 


: r. n n 





The chapters which go to make up this book are reprints, — almost verbatim, the 
editorial " we " even having been retained, — of articles which appeared in The Engineer 
during the present year. The first was published on February 14th and the last on 
August 29th. 

To the average individual, perhaps, electric welding appears to be an outcome of 
the War ; an expedient rendered necessary by the insistent demand for the speeding 
up of production by all available means. As a matter of fact various processes of 
welding by the aid of the electric current, notably carbon arc welding, have been in 
commercial operation for a quarter of a century or more, this country having been 
among the first to adopt and develop some of them. »Some of them, too, had, long 
before the War, been much in vogue both on the Continent and in the United States 
of America. In the case of the latter country electric welding had, for years before 
the outbreak of hostilities, been most successfully resorted to for many purposes to a 
considerable extent, especially for repair work on railway rolling stock. Sweden, 
among countries ou this side of the Atlantic, may be cited as being in the forefront 
in the employment of electric welding for ship repairing. 

Nevertheless it is quite true that the War gave such an impetus to the use of the 
newer method of welding as many years of endeavour during times of peace would have 
failed to impart to it. Moreover, the War greatly enlarged the sphere of its application, 
and was instrumental in bringing home to our manufacturers what a powerful agent 
was the electric current for the carrying out of certain types of work. As a consequence 
it has, nowadays, become the rule, rather than the exception, to find electric welding 
plants in our factories, and there appears to be every probability of such plants being 
even more extensively employed than they are at present. Furthermore, the War has 
undoubtedly had the effect of developing and improving the machines and appUances 
which both arc and resistance welding call for. 

The central idea around which the articles were written for The Engineer was to 
impart to the readers of that Journal a good general knowledge of the different 
systems of electric welding which have been evolved and of the various machines and 
pieces of apparatus which are required or are desirable for their efficient operation. 
No attempt was made to go, in detail, into the history of the subject, though certain 
historical matters are referred to. Nor was it sought to provide a handbook to the 
subject by giving minute" instructions as to the correct manner in which to work the 
various systems. To have done either, or both, would have entailed the introduction 
of a vast amount of matter which would have resulted in the enlargement of the 
volume to unwieldy dimensions, as well as in the production of a sense of weariness 
in the mind of the reader, who, if he intends to adopt electric welding, in one or other 
of its various forms, will naturally seek for practical instruction in the manner of 
working it, which it is, of course, impossible adequately to convey in a book. Again, 
it was considered undesirable to endeavour to enumerate, at length, the possible 


applications of electric welding. Such an effort would have proved to be a futile task 
since the number of such applications multiplies from day to day. Yet it will be 
found that mention has been made of such a varied selection of applications as will 
enable the reader to form a very fairly accurate conception of the wide area which has 
already been covered. 

It should be borne in mind that, though discovered in the middle eighties, and 
though usefully employed ever since that time, electric welding, as a science, has 
really, as yet, scarcely passed its infancy. It is very certain that the future — possibly 
the very near future — will witness great developments in connection with it. Never- 
theless, it does not appear to be likely that there will be any great fundamental 
departure from the general principles, either of arc or resistance welding, which it has 
been the endeavour of the author to deal with in this volume. 

LoNDOX, November, 1919. 



Introduction .............. I 

The Benardos Carbon Arc Process ......... 8 

Resistance Weirding ............ 14 

Arc Welding at a Steel Barrel Works . . - . . . . . .19 

The Pontelec Methods and Machines ......... 25 

The Quasi-Arc Process ............ 33 

Resistance Welders of the British Insulated and Helsby Cables, Ltd. ... 31) 

Machines and Apparatus for Arc Welding ........ 49 

Machines and Apparatus for Arc Welding icorUinued) ...... 58 

Machines and Apparatus for Arc Welding (continti^d) ...... 65 


(1) Resistance Welders of the Electric Welding Co., Ltd. .... 75 

(2) Some Large American ** Spot ** Welders ........ 78 

Oil-Drum-Making by Resistance Welding ........ 84 

Resistance Welders of the A 1 Manufacturing Co., Ltd. ..... 89 



Thr Strknoth i)F Ki.ncTBir Welds , . . . 

A Laror BRmsH " Spot "-Weldiso Machine 

The " Pijistic-Abc " Wkldinw System 

The " A.C " System or Arc Wrldiso 












Helmet for Abc Welding 

Examples of Welding (Benardos) 

The Benardos Electrode Holder . 

Device for ** Point *' Welding (Benardos) 

The "Electric Forge" (Benardos) . 

Device for Continuous or for *' Point " Welding (Benardos) 

Suggested vFormation of Armour Plate (Benardos) 

Diagram of Thomson Welding System 

Diagram Showing Principle of ** Spot '* Welding 

Diagram Showing Principle of " Seam " Welding 

Diagram Showing ** Butt " Welding of Tyres 

Joint Welded by Acetylene . . . 

Carbon Electrode Holder (T. T. Heaton) 

Welding-in the End of a Barrel . 

Putting on Stiffening Ring to a Barrel 

** Ponteleo " Resistance Welder 

Machine for Brazing Collars on Tubes (" Pontelec ") 

Spectacle-Frame Soldering Machine ("Pontelec'*) 

Typical *' Pontelec " ** Spot *' Welder 

** Spot " Welder for Bench Work (** Pontelec "} 

" Pontelec " " Butt " Welder .... 

Typical ** Pontelec " '* Seam " Welder . 
Diagram Illustrating " Quasi- Arc " Welding 
Welded Riveted Joint ...... 

B.I.W. " Butt " Welder, No. 57 ... . 

B.I.W. Chain-Link-Forming Machine 

B.LW. " Butt " Welder 

B.LW. Chain Welder ...... 


30. > Examples of B.I.W. "Seam" Welders 



32. Group of B.I.W. No. 13 "Spot" Welders 

33. Westinghouse Arc Welding Equipment on a Motor Lorry 

34. Diagram of Connections of Westinghouse Generator for Arc Welding 
36. Arrangement of Premier Electric Welding Co.'s Portable Welding Set 

36. Premier Welding Set in Cabin 

37. Portable Plant of the Premier Electric Welding Company 

38. Lancashire Dynamo and Motor Co.'s Arc Welding Set 

39. Connections of Lancashire Dynamo and Motor Co.'s Arc Welding Set 

40. The Davies-Soames (Daysohms) Electro-Magnetic Differential Clutch 

41. The' Davies-Soames Clutch Connecting a Motor to a Generator . 


















































Davies-Soame8 Clutch arranged for Arc Welding 

Diagram of Connections of Davies-Soames Clutch 

Electric Connections of " Constant-Energy " Balancer Set 

Holmes Portable Generating Set for Arc Welding 

Holmes Resistance Frame for Arc Welding . 

Tillino-Stevens Lorry equipped for Arc Welding 

Tilling-Stevens Arc Welding Lorry — Interior 

Tilling-Stevens Arc Welding Lorry — Driver's Seat 

Tillino-Stevens Arc Welding Lorry 

Crompton Self-Contained Arc Welding Set . 

Crompton Motor Generator Set for Arc Welding 

Equipment and Engineering Co.'s Electrode Holders 

Electric Welding Co. — ** Butt " Welder for Tubes, Rods, etc. 

Electric Welding Co. — " Butt " Welder with Radial Welding Blocks 

Electric Welding Co. — ** Butt " Welder for Miscellaneous Repair Work 

Electric Welding Co. — " Butt " Welder for Strips, etc. 

Electric Welding Co. — " Butt " Welder for Hoops, Rings, etc 

American " Spot " Welder for Ship Work 

Large American Duplex " Spot " Welder 

Electrical Arrangements of Dltlex Welder 

" A 1 " " Spot " Welder . 

** A 1 " Automatic Switch 

" A 1 " ** Butt " Welder . 

" A 1 " " Butt " Welding Attachment 

" A 1 " " Se^m " Welding Attachment 

Welded and Riveted Joints . 

Large *' Pontelec " ** Spot " Welder, with Stakes arranged Horizontally 

Large ** Pontelec " " Spot " Welder, with Stakes arranged Vertically 

Large ** Pontelec " " Spot " Welder — Upper and Lower Stakes or Arms 

Large " Pontelec " " Spot " Welder — Arrangement of Electrode Slide 

Large ** Pontelec " ** Spot '* Welder — Arrangement of Transformer 

S.S. Oeorge Washington — Fractured Circulating-Pump Casing 

S.S. Prim Joachim — Broken Cylinder-Head and Broken Low-Pressure Cylinder 

S.S. Kaiser Wilhelm II — Fractured Low-Pressure Cylinder Liner 

S.S. Friedrich der C/rcw«c— Broken Steam Intake, Second Intermediate Cylinder 

" Plastic- Arc " Welding and Cutting Panel ...... 

** I*lahtic-Arc " Welding and Cutting Panel, Electrical Connections . 
Cracked Loc()mc)Tive Frame — Early Stage of Repair by Arc Welding 
Cra<'Kki) Locomotive Frame -Final Appearance of Repair by Arc Welding 
The " A.('." Klkctiuc Welding and Cutting Machine .... 

Mkthodh ok ('onnecting up the *' A.C. " Machines with Different Types of 

' < I RO 1 ' I Tn .••*.•.... 

The **A.('." Electrode Holder ...... 

The ** A.C." Ele(;trodk Holder Gripped to Release Electrode 












































I ' 

I I 




Electric welding, which is far from being a novelty, has recently come prominently 
into public notice. It has not only been employed by the Admiralty for the con- 
struction of vessels, but Lloyd's Register has issued regulations under which it is 
prepared to sanction its use in shipbuilding operations generally. Moreover, the 
astonishing rapidity with which the damaged interned German liners were repaired 
by this process, thus enabling many thousands of American soldiers to be brought 
to Europe long before such transportation was deemed to be possible, is stiU fresh in 
the public memory. We feel sure, therefore, that a discussion of the subject will be of 
interest to our readers, and we propose in the present and succeeding chapters to 
deal with the question at some length. No attempt will be made to go in detail into 
the evolution of the various processes involved, our purpose being, rather, to show 
1 what has been and is being done, than to relate exactly who was the inventor of this 

and who was the discoverer of that. It will suffice to say that, quite early in the use 
of electric energy, experiments with a view to using the current for welding purposes 
were instituted, and that thirty years and more ago electric welding was practised 
notably by that prolific inventor and distinguished investigator, Elihu Thomson, and 
by the Russian Benardos. Of the work of these two men we shall have much to say, 
but we shall not attempt to deal with it exhaustively, so that what follows in no way 
claims to be a complete history of their discoveries, nor, indeed, of electric welding 
as a whole. It is noteworthy that though introduced at no widely diflEering periods of 
time the methods of Thomson and Benardos were entirely dissimilar, the one employing 
what is termed the resistance process and the other the arc process. All the electric 
welding systems in use at the present day may be classed under one or other of these 
two heads, though there are numerous variations in the application of the two 

Resistance welding resembles much more closely the original welding of the black- 
smith than does arc welding ; indeed, saving that the manner of arriving at welding 
heat is difEerent in the two cases, the welding of the gimithy is identical with the welding 
by the resistance method. In both the portions of the parts to be joined together 
are raised to a proper heat in the immediate neighbourhood of the proposed weld, 
and then welding is eflfected either by means of pressure or by percussion. The same 
precautions regarding the cleanness of the surfaces to be joined together have to be 
exercised in each case. Arc welding is essentially different in that the metal is in the 
majority of cases actually melted at the weld, and that neither pressure nor hammering 
is absolutely necessary, though in some cases, they may be beneficially applied. As 
we shall show, however, there are cases in which the arc is used simply to obtain 
welding heat. In such instances, of course, percussion or pressure is required to con- 
solidate the welds. Both arc and resistance welding are being more and more exten- 
sively employed and, from present indications, it would appear highly probable that 
both will continue in vogue, since each has its own sphere of usefulness, in which it 
performs with better effect than does the other. Nor does it seem likely that either 

1E.W. B 


will oust the acetylene method, which likewise iB^re-eminent in certain directions ; 
£0 that the engineer of the present day is fortunate in having three more methods by 
which welding can be effected than had his predecessor of the last generation. It may 
be even said that he has more than that if the oxy-hydrogen or oxy-coal.gas methods 
be taken into account. 

In both resistance and arc welding a certain amount of special machinery or ap- 
paratus is required, though with the latter it is generally less elaborate than with the 
former. For resistance welding alternating current is almost universally employed 
in practice, though there is no reason why direct current should not be used, saving 
the difficulty of dealing with the heavy currents. In arc welding, on the other hand, 
either direct or alternating current can, under certain circumstances, be employed, 
though direct current is most generally used. Resistance welding may be effected by 
three methods : — (a) Butt welding ; (6) spot welding ; and (c) line or seam welding, 
together with several modifications. Arc welding may be performed either with bare 
metal electrodes, with flux-covered metal electrodes, or with carbon electrodes. 

The quality of the current — ^if such an expression be permitted — employed in the 
two processes differs very widely. In resistance welding there must always be a very 
heavy current — in some cases it amounts to thousands of amperes — and a very low 
pressure, say, from J to 6 volts. Some largo machines have recently been built in 
America in which voltages considerably in excess of the higher of these figures have 
been employed. We shall refer to them later. In arc welding, on the other hand, the 
voltage ^must always be high enough to maintain an arc — say, from 20 to 56 volts — 
and the current is relatively small. As a matter of fact, it is found in practice that the 
electromotive force available should, at any rate with metal electrodes, be at least 
from 60 to 75 volts, while some processes require an available pressure of from 100 to 
110 volts.* With carbon electrodes a pressure of 90 volts is commonly employed. 
The current in arc welding is adapted to the nature of the work being carried out. 
It may be quite small, say, from 15 to 20 amperes, and it is very rarely, we believe, 
that it exceeds 800 amperes with carbon electrodes, and less than half that with metal 
electrodes, the average in actual practice being considerably less than the smaller 
figure. In the arc process, as originally introduced by Benardos, the work which was 
to be welded was connected to the negative terminal of a source of electrical energy — 
in the first instance storage batteries were employed — while the positive terminal was 
connected to the carbon electrode. An arc was formed by touching the work with the 
electrode and quickly withdrawing the latter, the result being that the metal im- 
mediately adjacent to the point of impingement of the arc was quickly raised to a very 
high temperature, and a stick of similar metal could also be melted just over the spot 
much in the same manner as solder is melted with a soldering iron. Later, this arrange- 
ment was found to possess the disadvantage that, as the direction of the current was 
from the electrode to the work, there was a tendency for the carbon, either in the form 
of vapour or in minute highly heated particles, to be carried over to the latter, so that 
it entered into combination with the molten metal, and sometimes, undesirably 
altered its composition. Hence the polarity of the work and electrode were inter- 
changed, the former becoming the positive and the latter the negative. A further 
reason is put forward by the advocates of the carbon arc process as to why the work, 
and not the metal electrode, should be made the positive terminal, and that is that 
the positive terminal of an electric arc is hotter than the negative terminal. We 
believe we are right in stating that carbon arc welding was first employed commer- 

♦ It will be found in a subsequent chapter that this statement is modified. The tendency would appear 
to be to reduce the voltage. 



ciaUy in this country by the firm of Uoyd and Uoyd, of Birmingham, which is now 
merged in the firm of Stewarts and Lloyd. It has also been used with success, for twenty 
years or more, by the Steel Barrel Co., Limited, of Uxbridge. 

Some ten years after this introduction of the Benardos process the idea occurred 
to another Russian — Slavianoff — ^to employ a metal electrode instead of a stick of 
carbon, his idea being to fill up blowholes in defective castings. Previously Benardos 
had suggested the employment of a cored carbon, the core consisting of metal. Whether 
such composite electrodes had any great vogue' we are unable to say, but having re- 
gard to latter-day developments in coated electrodes the suggestion coming, as it 
did, as far back as the year 1885, is distinctly interesting. Covered metal rods are, as 
we shall show later, in very extensive use at the present day, but the covering is not 
carbon ; indeed it is, for the most part, a non-conductor. It is interesting to note 
that in 189*3 — some two years before Slavianoff introduced the use of metallic elec- 
trodes — Mr. T. T. Heaton, of the Steel Barrel Company, Limited, had suggested a 
mild steel electrode to the late Mr. Dickinson, of the Bowling Iron Company, and 
that it was tried, though without success, the experiment not being persevered with. 
The Slavianoff method ha^, of recent years, been very considerably developed, and 
though it would be erroneous to imagine that the metal electrode has entirely displaced 
the carbon electrode, for it has done nothing of the kind, yet it is safe, we beUeve, 
to say that in the large majority of welding installations in which the arc process is 
employed, metal electrodes of one kind or another are used. 

A development of the carbon arc process is that in which the arc is struck between 
two carbon rods arranged at an angle to one another, the angle pointing downwards. 
By means of an electro-magnet arranged between the carbons, the arc is projected 
downwards on to the work. The device is cumbersome and not particularly easy to 
' ^PP^y> ^0 ^^^^ ^^ ^^ hardly surprising that it has not come largely into use, though there 
would appear to be no reason why it should not do its work perfectly well. In the only 
example we have seen, the arrangement was such as we have described above, and the 
whole apparatus was suspended from a hook. It was not, therefore, possible to use 
it for vertical or overhead work. Its invention is attributed to Zerener, of Leipzig, 
and, to the best of our knowledge, the system is only in operation in Germany, and 
only there to a quite limited extent. 

For arc welding the special appliances actually necessary are but few. In addition 
to the source of electric current and the electrode with its holder, all that is essential 
is a steadying resistance, connected in series with the arc, and possibly an ammeter 
as weU. As will be seen later, however, Special machines have been introduced with 
the purpose of preventing rushes of current. It will be readily understood that, unless 
means for controlling the current are employed, a resistance, or some such device, 
in series with the arc is an absolute necessity, for without it when the electrode touches 
the work there is a complete short circuit until the arc is struck. Although, however, 
but few special appliances are required, it is generally found that the supply of current 
available is not suitable for welding. For instance, it is very rare nowadays to find 
a pubUc supply with as low a voltage as even 1 10. Hence it is necessary,, if the current 
be not specially generated, to reduce the pressure so that there may not be an 
excessive waste of energy in the resistance. If the current be alternating the reduc- 
tion in pressure is readily effected by means of a static transformer, which may have 
different tappings in the primary, so that different voltages and currents to suit 
different kinds of work may be obtained. This arrangement, of course, applies only 
if alternating current is to be employed in the arc. As a matter of fact, direct current 
is, as we have said, almost universally used, at any rate in this country, and to pro- 


duce the low voltage direct current a motor generator may be employed. The 
direct cmrent half of motor generators designed for arc welding are ordinarily com- 
pound wound, and are preferably provided with commutating poles, so that sparking 
may be minimised. The size of machine required depends, naturally, upon the 
nature and amount of the Work being carried out. The variation in the. current 
when using metallic electrodes may be between, gay, 15 and 250. These may be 
regarded as extremes, only met with in special cases, the more usual range 
experienced in actual practice being between, say, 20 and 200 amperes. With 
carbon electrodes, on the other hand, much heavier currents are experienced. We 
understand that satisfactory results cannot be looked for with currents under 
300 ampdres, and that not infrequently 500 and in special cases as many as 
800 amperes may be required. With these figures an estimate may be arrived 
at regarding the capacity of the machine necessary for any given number of 

At this point, perhaps, a few words may be said with regard to the arc itself and 
its eflfects on the human frame. In the first place, the arc with a carbon electrode is 
long and the temperature in the positive crater is very high. In a paper which he read 
.before the Institution of Engineers and Shipbuilders, in Scotland, on June the 22nd, 
1918, Captain James Caldwell, R.E., Deputy Assistant Director of Materials and 
Priority, stated that it was estimated to be 7600 deg. Fah. It is not, a6 Captain Cald- 
well explains, necessary to reach such a temperature as this in welding, and the carbon 
rod is therefore moved along at a rate which will ensure good furion to the necessary 
depth. With metal electrodes, partly because of the shorter arc and partly because 
of the pressure of the vapour of the metal, it is lower but still high. In both cases, and 
particularly in the former, ultra-violet rays are freely emitted. The eifect of exposing 
the eyes, face, and hands to direct rays from the arc is very similar to severe sunburn, 
such as is familiar to climbers in high altitudes on freshly fallen snow. The results 
of being burnt in this manner by careless exposure are exceedingly painful, and 
whereas there are those who say that no permanent injury is caused to the eyes of 
operatives constantly engaged in electric welding, there are also those who aver that 
cataract may result in cases of careless exposure. However, with reasonable care 
and with proper precautions, there need not, apparently, be any ill-effects whatever. 
It is necessary, however, adequately to shield not only the eyes but the face and neck 
as well, and, of course, the hands. In some cases screens sufficiently large to protect 
both the head and neck are employed, the screens being furnished with glasses of colours 
of such a character that while the harmful rays are cut off the operator can readily 
see what he is doing, at any rate when the welding is actually in progress. Then, too, 
aluminium helmets, or cylinders, put on over the head and resting on the shoulders, 
are used. An example of a helmet which is being employed in some shipyards is shown 
in Fig. 1. It will be observed that the protective glasses can be raised when required 
so as to give the operator unobscured vision. It is customary also for heavy leather 
gauntlets to be worn so that both hands and wrists may be protected, and, though 
ordinarily the clothing is sufficiently thick to give protection to the body, it is the 
wisest course to wear a leather apron, reaching down to the boots, for splashes of metal, 
and, when coated electrodes are used, of molten slag, may be projected from the arc. 
Coat-sleeves must, in no case, be rolled up so as to expose the forearms. These precau- 
tions may sound formidable, but in reality they are not so, and in order to ensure 
immunity from injury, which may be exceedingly painful if it be nothing worse, it 
is imperative to take them. Further than this, it is most desirable that any workshop 
or compartment in which arc welding is in progress should be adequately ventilated. 


The fumes somotiniea given off are offensive, and if not carried away as fonned may 
injure the health of the operatives. 

It is important, too, that the colours of the glass screens should be such that they 
entirely absorb all the ultra-violet rays. Certain combinations of blue and deep red 
glass will effect this absorption. So will glass having a greenish amber tint, and this 
colour is, we gather, largely used in the United 8tat«s. There ore other glasses, too, 
for which it ia claimed that they are impervious to ultra-violet rays. We shall have 
more to say regarding this matter in a later chapter. 

There is a great deal of difference of opinion as to whether or not in arc welding 
a flux IS necessary. There are those who allege that, whereas it is not required with 
metal electrodes, it is most desirable when carbon electrodes are used, since it enters 
into combination with any carbon which may be projected in the direction of the 

Fio. 1. — Helmet for Arc Welding, 

weld and prevents it combining with the metal. Then, again, there are those who 
maintain that a flux is necessary to increase the fluidity of the metal. It is urged, 
too, that welding by means of the carbon arc without a flux results in the fonnation 
of a crater of boiling metal, which greedily absorbs oxygen from the atmosphere, and 
that a weld which has absorbed oxygen will always be weak, because the oxygen will 
react and combine with the carbon in the metal, and with the metal itself, and the 
metal becomes poroug. The same allegations are made regarding the use of metal 
electrodes without flux. Into this controversy we do not, at the moment, propose to 
enter ; nor do we propose to discuss the relative merits of bare and coated electrodes. 
Each has its own advocates. In the United States bare electrodes were, as we have said, 
almost exclusively employed up to within quite a short time ago ; whereas, in this 
country covered metal electrodes have been most extensively used. In both countries 
a large amount of excellent work has been carried out, and we are not prepared, with 
our present knowledge, to adjudicate between the two methods. It is significant, 
however, that in the case of the electrically welded bai^e, to which we made reference 
in our issue of August 9th 19IS, and work on which was carried out under the supw- 


intendence of the Admiralty authorities, and under the eyes of experts from lioyd^s, 
metal electrodes, provided with a special slag-forming coating, were exclusively used. 
It is noteworthy, too, that coated electrodes are much more frequently used in the 
United States nowadays than they used to be. It is also an undoubted fact that 
excellent work can be and is constantly being done with the carbon arc on mild steel 
without the use of any flux whatever. The question is rendered all the more difficult 
i^ that there is no sort of agreement as to the composition of the flux or the flux- 
forming coating which it is best to employ, and because an excessive current in the 
arc may bring about a condition in the weld which is very similar to, if not actually 
identical with, that stated to be caused by the use of, say, a carbon electrode without 
a flux. The only point on which there seems to be no divided opinion is that if a flux 
is to be used, the most convenient method of applying it is to coat the electrode evenly 
with it. One authority on the subject states, with regard to this point : "When 
appUed in this way the coating may be made to serve two useful purposes, providing 
it contains sufficient refractory material : (1) It permits the electro-magnetic effect 
of the welding current to act upon the molten metals ; and (2) it appUes flux to the 
weld at the rate that adjusts itself perfectly to that of the melting of the welding 

It should be mentioned, however, that all the coatings applied to metal electrodes 
are not flux forming, or, we should more correctly say, Uquid flux forming. In what 
is known as the "gaseous flux process,'* which was introduced and patented by 
Kjellberg, the metal electrode is — ^again to quote Captain Caldwell — " covered with 
a fireproof sleeve of non-conducting material, so that as the metal is removed by the 
arc the sleeve projects beyond the end of the rod and forms a guide for the molten 
welding metal, the sleeve itself falling away automatically. The sleeve protects the 
metal from oxidation and reduces heat losses." ** The process," continues Captain 
Cakiwell, " is an improvement on the bare metal electrode, and a great deal of satis- 
factory work has been done with it, mainly in repairing marine boilers, stem frames, 
and other ship parts. It is also successfully used to build up worn parts, such as 
propeller shafts, worn crank shafts and axles, which are afterwards machined to 
size. The patent specification claims the use of no particular material for the fire- 
proof sleeve, and no other purposes than those mentioned, but the company ex- 
ploiting the process states that the sleeve can be made the vehicle for constituents 
which will give desired characteristics to the added metal." The arc actually dissi- 
pates the flux, so that no slag is formed. 

It is not to be supposed that electric welding will always produce work which is free 
from defects, for that is not the case, and unfortunately, it is by no means always that 
a defect, should it exist, can be detected, even by careful observation from the outside. 
In resistance welding some foreign matter may have got between the surfaces being 
operated upon and have reduced the area properly welded. In metal arc welding, 
if the operator has used too smaU a current, or held the electrode too far away from 
the work, the melting of the edges of the work may not have been properly effected, 
so that there is not adequate adhesion between them and the added metal. The joint, 
to outside appearance, may look all right, but there may be cracks between the work 
and the added metal which will, in time, certainly lead to trouble. An excess of current, 
on the other hand, will have the effect of what is known as " bad metal " being de- 
posited. Then, again, bubbles of gas may form in the weld, and hence reduce its 
strength. It is quite possible, too, when using a carbon arc incorrectly, to burn the 
metal by using too heavy a current or by keeping the arc playing on one particular 
spot for too long a time. However, it is by no means always that a perfect wel<J is 


obtained, either by the blacksmith or by the use of the acetylene torch, and it is prob- 
ably accurate to say that, with arc welding at any rate, the proportion of imperfect 
welds on any given piece of work made by even a comparatively unskilled worker, is 
as small as, if it be not smaller, than is the case with any other welding process. More- 
over, there is no reason to suppose that when reasonable care has been taken resist- 
ance welding produces — ^in general practice — ^welds which are inferior to those 
obtained by the smith, and it certainly can do work which he cannot carry out. 

We shall have more to say regarding the strength of electric welds in a later 
Chapter.* I 

♦ See Chapter XIV. 



As we said in our first chapter, this series is by uo means intended to constitute a 
history of electric welding. To have attempted to make it do so would have intro- 
duced an immensity of weary details, and made the whole thing unwieldy. Hence, 
we shall not discuss the claim made for Benardos that he invented arc welding, further 
than to say that there are those who allege that de Meritens, whose pupil Benardos 
was, was the actual discoverer of the arc welding process. On this point we shall 
not make any definite assertion ; nor is it neoessary for our present purpose to do so. 
It can be truthfully said, however, that the Russian was largely instnimental in 
bringing the subject to pubUc notice, and that in 1885 he, in conjunction with a fellow 
countryman of his, one Stanilaus Olszewski, an engineer of St. Petersburg, obtained 
a British patent — ^No. 12,984 of that year — ^in which the first claim read as follows : 
" The method substantially ad herein described of uniting and soldering metals together, 
either at separate points or in a continuous joint by heating them by means of an 
electric current," and that the methods described embodied the employment of the 
carbon arc. It would be interesting to inquire into the question of how much of the 
invention was due to the one inventor and how much to the other. Nicholas de 
Benardos is described in the patent specification as ^' Gentleman," his partner, as 
we have said, as an engineer. What part did he, whose name has alone survived as 
being coupled with the invention, actually play in the discovery ? This, again, 
is a point which we shall leave to others to discuss, and shall confiine ourselves to giving 
certain extracts from the specification, which is certainly a most comprehensive 

The invention is described at the outset as relating to ^'an improved method of 
and apparatus for the direct application of electric currents for the following purposes : 

1. The union of metals. 

2. Their disunion or separation. 

3. The formation of apertures in metals. . 

4. The union of metals in layers. 

The process, which we term eleotrohephaest, consists of the formation of voltaic arcs 
when necessary." It is explained that the voltaic arc is formed by the approach 
of carbon — or any other body sertnng the purpose, the italics are ours — to the part of 
the metal to be operated upon, the carbon forming one pole of the circuit and the 
metal the other pole. 

It is curious what wide area of ground is covered and how many are the uses to 
which the inventors claim that their invention can be put. It is pointed out in the 
first place that the union of parts of one or more kinds of metal may be effected in two 
ways, namely, "either in points or continuously: 1. In points by agitating the 
voltaic arc at one point for a certain time, after which it ceases ; 2, continuously, 
in which case the voltaic arc advances uninterruptedly in a defined line. In both 


cases, the union or cohesion of the parts is simultaneous, and they become one homo- 
geneous whole." One is at once struck with the similarity between the processes 
described and present-day spot and seam welding, though nowadays spot welding, 
at any rate, is almost invariably, if not universally efiEscted by the resistance process, 
and many kinds of seam welding are only practicable by the latter method. " Point 
union," the inventors declare to be " in svery way superior to riveting," while " Con- 
tinuous union " is also described as superseding riveting, "particularly in the case 
of boiler joints, etc., and is generally applicable in all cases of joining or uniting metals." 

A variety of different methods of making joints in iUustrated. Among them is a 
joint formed between two plates by welding the slightly projecting ends of headless 
rivets, and another in which only one of the plates has holes drilled in it, and the joint 
is effected by filling up the holes with , 
melted metal which itself becomes 
welded to the metal of the imperforated 
sheet. Sketches roughly illustrating 
these two methods of jointing are 
shown in Fig. 2. Presumably, hammer- 
ing was re&orted to in order to con- 
solidate the weld, as is indicated in the ' "" """"pi^^ 2.-Ei«mple8 ot Wdding. 
right-hand joint in each case, but 

hammering is not actually mentioned in connection with joints of this type, though 
the utility of hammering or pressing was fully understood by the inventors, as we 
shall show later. 

The inventors were not backward in praising their process, as is amply evident 
from the following extract, which is taken from that part of the specification which 
immediately follows the reference to jointing : " As compared with existing processes, 
the advantages claimed for the present invention are its rapid action, cheapness, the 
reduction in weight owing to the absence of joints, cover joints and heads of rivets. 
By this process the strength of the joints becomes equal to that of other parts of the 
objecf, which is most important in boiler and armour plate, iron and steel vessels, 
hydrauhc apparatus, etc., and had hitherto been attained with great difficulty. Besides 
the advantages above enumerated, the great feature of our invention consists in the 



If _______ 

Fra. a— The Benardos Electrode Holder. 

facility with which it can be applied on the spot in cases of emergency or accident 
without taking the object to pieces : as, for instance, in the case of the bursting of 
boiler plates, the cracking of armour plates, etc." 

The carbon electrodes which might be used are referred to as being of various forms, 



the mo8t practical for operations on a small scale being ^' either a solid or hollow pencil 
or rod. The hollow carbon is filled "with various metals or their alloys, which play the 
part of solder." To this we shall refer again later. Of holders, various forms are 
LUustrated. The simplest is very similar to that in u^e at the present time. It is shown 
in Fig. 3. The hollow at the end of the wooden handle contains a contact screw which 

Fig. 4.— Device for "Point" Welding. 

serves to fix the wire of the conductor, while the electrode consists of a carbon pencil 
held in a tube by means of a screw. This carbon-containing tube works on a hinge, and 
may at the will of the operator assume ^^arious inclinations with regard to the lever 
which connects the conductor with the carbon. A more complicated arrangement. 

Swain Sc 

Fig. 5. —The "Electric Forge." 

which is specially intended for use in " point " welding, is shown in Fig. 4, which 
gives a cide view of the apparatus. It is, as will be observed, furnished with a, lever 
or trigger for pressing the carbon down to the point where the arc is to be formed, this 
motion being accomplished by simply pressing the lever. The breaking of the arc is 


effected when tlie pressure on the trigger ia removed by the action of a spring pressing 
against the trigger, tending to force it downwards. The conductor is fixed to a contact 
post contained in the handle of the lever. A still more complicated device is shown in 
Fig. 6. It is intended for use either in " point " or continuous welding, and is furnished 
with a protecting screen and also a stop watch, which was started by the some 
mechanism which struck the arc, so that the observer might see how long the current 
had been at work. When it was to be used for continuous welding, the rails, on which 
the apparatus ran on wheels, were to bo of the ordinary plain topped type, but when 
employed for "point " welding one of the rails was to be corrugated, the distance 
between the corrugations determining the distance apart of the welds. 

Reference may also be made to Fig. 5, which shows what the patentees called an 
" Electric foi^e." It consists of an anvil carrying a bracket, furnished with a pulley, 

Fid. 6. — Device for Continuous or for "Point" Welding. 

to which is attached a pair of tongs for holding the object to be operated upon. The 
carbon is arranged on a stand to which is fastened one conductor leading to the source 
of current. The other conductor, which is attached to the anvil, is in electrical contact 
with the object to be heated. When the latter is touched on the carbon electrode and 
withdrawn, an arc is formed, the object is heated and is then transferred to the anvil 
for forging. 

By their process the inventors claimed that fusion of layers of metal one on top 
of the other might be brought about by introducing into the arc a " carbon pencil 
prepared with the metal with which the fusion is to be effected, which falls drop by 
drop in a continuous jet. The pencil is insulated. . . . This process can be employed 
in a number of cases for soldering, for filling up flaws, cracks, hollows, etc., moreover it 
can be applied for steelifying objects for makii^; stayB, supports, etc." By steehfying 
the inventors intended to convey, we take it, hardening or tempering. The use of the 
arc for making holes is claimed and so is " the separation and disunion of metal," or 
as we call it nowadays, simply " cutting." " Farts of metal of considerable dimensions 



may," says the specification, "be detached by this process with facility on the spot. 
The tension and quantity of the electric current must be increased in proportion to 
the size of the metal operated upon. Numerous cases occur where the separation or 
detachment of metallic part^ cannot be effected by eidsting means, and which could 
be easily accomplished by our process." 

One of the most curious claims made for the process was, however, its use for the 
formation of armour plates. The resulting product is shown in Fig. 7. The steel 
plates were to be joined together both by point and continuous union. They were to be 
laid one on the top of the other, with intermediate layers of some elastic substance 
such, for instance, as caoutchouc, felt, cardboard, wood, etc. The elastic substance 
is shown in the drawing by shading. The inventors point out that these armour 
plates might serve for protecting vesselsl and for numerous other purposes ; that they 
are exceedingly light, offer great resistance, are not hable to crack and split, and that 
they may, with great facility, be made on the spot where they are to be used, either 
whole or attached layer by layer when they are being mounted. 

The inventors evidently realised that in some of the processes carried out in 











\/^j^ji^j ^ ^^\ 























The Enqimeer" 

FiQ. 7. — Suggested formation of Armour Plate. 

Swain Sa 

accordance with their invention, consolidation of the welds would be necessary, 
for they state that the apparatus described " can also be directly connected with'other 
instruments, as, for instance, with the rolls, stamping press and automatic hammer 
to which the metal is automatically conducted on the cessation of the arc, for the 
purpose of further cohesion and final planning." They were also aware of the necessity 
for protecting the eyes of the user, for they explain that the various pieces of apparatus 
described are fitted with " Coloured glass screens or some similar substance, so as to 
allow of the operation being watched without injury to the eyesight of the operator. 

We give one further extract from this interesting document : " This process of 
working metals may be combined with a gas apparatus, whereby gas or a mixture of 
gases could be introduced by pipes into the voltaic arc, and increase its tension. Into 
these gases may be introduced pulverised carbon, which is a good conductor of elec- 
tricity, and also increases the action of the voltaic arc. If it be necessary to fuse any 
heterogeneous substance with the metal operated upon, the carbon may be replaced 
by any other substance in powder." 

It must be admitted that this specification, the date of which, it should be remem- 
bered, is 1885, very nearly covers the whole sphere of electric welding as it is now 
practised. Butt welding, spot welding, seam welding, and the building-up process 


are all of them either mentioned by name or shown in the illustrations. It is true that 
the specification had to be amended in 1891, but none of the points which were excised 
in the emendation, have been referred to in the foregoing. It is quite clear that it was 
foreseen that electric welding was bound to assume a very important position in engin- 
eering construction, and it is a matter for wonder that, though in several outstanding 
examples the process has been successfully employed commercially for getting on for a 
quarter of a century, it is only comparatively recently that it has in any sense been 
widely used. It is not so very long ago that the Admiralty refused to permit the use 
of electrically welded tanks, though it has now definitely relinquished that position. 
Moreover, as showing the trend of feeling in other directions,* we may say that we 
have before us a copy of the railway companies' specification for iron or steel drums, 
suitable for the conveyance of acetone, acetone oils, ketone oils, coal tar, naphtha, 
benzole, toluol, turpentine substitutes, xytol, mineral naphtha, petroleum, benzine, 
benzoline, carburine, motor spirit and petrol, in which it is set out that not only 
must all the joints be electrically welded, but that each drum must also be provided 
with a well-fitting screw bung, the boss of which must be electrically welded to the 
head of the drum. 



Resistance welding, like many other valuable inventions, was discovered by accident. 
As it happened, too, the discovery of the process came at a most opportune time, for 
early makers of dynamos were seriously handicapped by the fact that the wire manu- 
facturers could not or would not supply the insulated wire in sufficient lengths. It 
was impossible to wind even a comparatively small coil without having to make one 
or more joints in it. These joints had to be brazed or silver soldered by means of a 
blow-pipe and then insulated by taping-up, with the result that the wire was con- 
siderably thicker at the joints than elsewhere. The extra thickness made the winding 
of a really neat coil a. matter of great difficulty, and the wire makers were urged again 
and again to make the joints in the first place before applying the insulation. Among 
those who endeavoured unsuccessfully to get them to do so was Professor Elihu 
Thomson, who was then (about the year 1885) connected with the Thomson-Houston 
Electric Company, of Lynn. The reply he got was that what he suggested was im- 
possible by any method known at the time. However, he himself was destined to 
discover a means by which the joints could be easily and efficiently made, and the 
story of how he came to do so was told in the following words by Mr. J. B. Clapper, 
plant engineer of the Rim and Tube Division of the Standard Parts Company, 
Cleveland, Ohio, in a paper which he read before the Cleveland Engineering Society, 
on March 26th, 1918. 

" While giving a lecture on electricity in general at the Franklin Institute of 
Philadelphia, one evening, among other apparatus for demonstration purposes, Prof. 
Thomson had with him a large high-tension jump spark coil and also some Leyden 
jars. After showing how a long spark could be produced from the secondary terminals 
of this coil by applying an interrupted direct current at the primary terminals, he 
then thought of trying an experiment with this coil the other way around. Accordingly, 
he arranged for one of his assistants to discharge several of the Leyden jars in series — 
thereby effecting a very high potential — through the secondary winding of the spark 
coil, while he held two wires together, end to end, which were connected across the 
primary terminals of the coil. After discharging the jars in this manner, the Professor 
found that he could not pull the wires apart that he held — ^they had become joined 
at the ends through the heating effect of the current passed through them — in short, 
the first electric weld had been made quite unintentionally. However, since the wires 
that the Professor held in his hands were copper, it immediately dawned on him — 
here was his solution to joining two lengths of copper wire together— electric welding." 

The Professor, being a business man as well as a distinguished scientist, at once 
decided to put his invention on a commercial footing. He patented the invention, 
and the first practical machine which he designed was known as the " Jews' Harp 
Welder." It was the parent of the very many different machines which are now 
employed for applying resistance welding for a large number of different purposes. 
In it, according to Mr. Clapper, the secondary winding was solid and of U form, the 
metal being thinned in cross section at the bottom of the U so that the ends on which 


the clamps for holding the ends of the wire to, be welded were mounted could be 
moved together or ap&rt enough to permit pushing up the stock in welding. The 
primary winding was a circular coil of many turns, laid within the secondary, which 
is said to have exactly resembled a Jews' harp in shape. The core consisted of many 
turns of iron wire wound '^ around and through the primary coil, embracing the 
secondary for its portion immediately adjacent to the primary winding." The pressure 
on the work was effected by a spring, tending to draw the outer ends of the secondary 
together, the tension being varied by a hand-operated screw passing through the 
spring. In 1888 the Thomson Electric Welding Company was formed commerciaUy 
to develop, manufacture and put on the market apparatus for adapting the Thomson 
process of welding to all lines of business coming within its scope. 

The original application of resistance welding was thus to the joining together of 
lengths of copper wire. It very soon, however, came to be used not only for iron and 
steel, but also for brass, aluminium and the finer metals, and how wide is the range 
of its applications nowadays may be gauged by the fact that the machines employed 
to carry it out vary in weight from considerably under half a hundredweight to 
many tons. 

As has been already mentioned, resistance welding resembles in several respects 
the welding of the ordinary smith. But, though like it in some ways, it is very different 
from it in others, and perhaps, before all, in that the heating and welding operations 
are performed practically at the same time and almost instantaneously. As soon as 
welding heat is reached, the welding is immediately effected, and in the lighter kinds 
of work the rapidity with which the welding heat is attained is quite remarkable. 
Although the process is not applicable to every sort and kind of welding, yet the 
sphere of its utility is very wide, and the quality of the work effected by it unquestion- 
ably good. It is necessary for its effective operation to have at command heavy flows of 
current, but, on the other hand, the voltage necessary is very low. In some cases as 
low an electromotive force as- half a volt is aU that is required. In actual practice, 
from four to six volts is about the highest pressure worked with, though in some 
special machines recently built in America higher voltages than those named are 
employed. The temperature of the metal to be welded is raised simply by allowing 
a very heavy current to flow through a restricted area, and to those who are unac- 
quainted with the process, the rapidity with which welding heat is arrived at when 
using properly designed and proportioned machines will appear extraordinary. 

As can be well imagined, the difiSculties attending the use of direct current for this 
system of welding are practically insuperable, except for the welding of exceedingly 
small articles. It is, however, only the difficulty of using it which precludes its com- 
mercial use in thid direction. It is not because resistance welds cannot be made with 
direct current. For the purpose, direct current would be just as effective as alter- 
nating, bub there is no comparison between the relative ease with which the two 
kinds of current may be applied. With the former, the heavy current has to traverse 
the whole of the machinery involved, whereas in an alternating system the heavy 
welding current only flows in part of 'the apparatus where it can be accommodated 
without any trouble. The supply voltage may be anything within reason. All that 
is necessary is co to proportion the primary and secondary winding of the welding 
machine that the desired secondary voltage is obtained. In the large majority of 
cases the secondary winding only consists, as in the original Thomson machine, of one 
turn, the two terminals of which are short-circuited by the object or objects which it is 
desired to weld. In such a case, supposing a secondary pressure of one volt to be 
required with a primary voltage of, say, 500, then the primary winding would be 


composed of 500 turns, and the conductor of the secondary irould need to have a 
carrying capacity of 500 times that of the primary. As for some purposes many 
thousands of amperes are required, it can readily be seen that the cross-sectional area 
of the secondary conductor must, in some cases, be very large. Generally speaking, the 
one turn of the secondary is built up of a large number of thin sheets or ribbons of 
hard copper joined in parallel, the reason for this subdivision being that flexibility is 
desired. The two sets of ends of these sheets of copper are 
attached respectively to two heavy copper jaws or terminal 
pieces, which have to be capable of motion towards and away 
from one another, so that the object or objects to be welded 
may be pressed together. The junction between the many 
sheetB of the secondary winding and these terminal pieces or 
jaws is effected as carefully as possible, so that there may be 
perfect electrical connection, and that each strand of the coil 
may carry its proportion of current. The jaws or terminal 
pieces are made of many fcrms to suit different kinds of 
work, and since they are liable to become very hot during 
continuous welding operations, they are, as a rule, made very 
massive, and are almost universally cooled by water circulated 
through cavities formed in their interiors. The simplest 
form of rcBistance welding machine is shown in Fig. 8, which 
represents dii^ammatic^y the Thomson principle. It will be 
realised that, if the articles or parts to be welded are introduced 
between the two terminals of the secondary winding, the circuit 
of the latter will be closed and current will flow if the primary winding be energised. 
Stated briefly, the machine is completed (a) by means for bringing the two terminals 
nearer to one another, so as Srst of all to close the electric circuit and afterwards, 
when welding heat is attained, to apply pressure to the heated parts, so as to form the 
weld, and (b) by means for energising and de-energising the primary winding. In, 
many cases these two motions are interlocked 60 that they may work in unison and 




Fia. 8.— Diagram of 
Thomson Wei ding System. 

FiQ. 9. — Diagram showing Principle ol Spot Welding. 

«o that the current may be cut off when welding beat is reached, and just before the 
final pressure comes on. The mechanism used for advancing and withdrawing the 
jaws and terminal pieces or electrodes is of a varjdng character, depending on the 
nature of the work to be dealt with, and on the taste and fancy of the designer. There 
are machines which are operated by a hand lever ; others in which the lever is worked 
by foot, while in some cases the motion is provided by a hand wheel and quick pitched 
Ecrew, and in others by a hydraulic cylinder. There are machines which are power 



driven, so that the top electrode is continuously being moved up and down, a weld 
being made at each stroke, if there be material to be welded between the jaiws or 
electrodes. Such machines may be automatic, in that the currently turned on and off 
and the pressure applied and relieved without requiring to be touched by the atten- 
dant, who has only to concentrate his attention on the articles to be welded. 



\ y Cilinder being Welded. 

( C \-Roll6r. 




Swain 8& 

Fig. 10. — Diagram showing Principle of Seam Welding. 

Some machines may be made available for a large number of different welding 
operations, but it is frequently necessary to have specially designed machines for 
special classes of welds. In fact, for each of the three resistance welding methods to 
which we have already made reference, i.e. (1) Butt welding, (2) spot welding, 

Jaws. ^ 

^Hg ENQiNccft" Swain Sc- 

Fig. 11. — Diagram showing Butt Welding of Tyres. 

(3) seam welding, many different types of machine have been evolved. To some of 
them we shall refer in later chapters. In aU of them, however, the electric principle 
is the same. Comparatively high voltage alternating current of smaU volume is trans- 
formed in a static transformer, which actually forms an integral part of the welding 

E.W. C 


machine, to large volume with very low voltage. The differraiees between the 
varioofl machines consist, among other things, in (a) size, (6) the means of applying 
the pressure and cutting the current on and off, and (c) in the forms of the 
terminal pieces, jaws or electrodes of the secondary windings, and of the arms 
carrying them. Just as in a hydrauUc riveting machine it is necessary- to have 
a considerable depth of gap, so it is with certain types of electric welders. The 
arms holding the electrodes are, in such cases, long. A diagrammatic example of 
such a case is shown in Fig. 9. On the other hand, there are many things which can 
be welded while held in the hand simply by being placed between two advancing 
jaws. In other cases, such as in that of such things as the rims of bicycles, the two 
ends to be welded have to be held tightly in clamps, which, when moved towards one 
another, cause the edges to be welded to press against each other. A diagrammatic 
sketch of such an arrangement is shown in Fig. 11. Then again, for seam welding the 
electrodes may consist of wheels or rollers as shown diagrammaticaUy in Fig. 10. 
Machines of this type are employed for the seam welding of cylindrical articles. 



The work carried out by the Steel Barrel Company, of Uxbridge, affords an excellent 
example of the employment of different welding processes for varying kinds of work, 
and would appear to confirm the assumption thab, as far as can be Seen at present, 
both arc and resistance welding, as well as the acetylene method, have come to stop* 
For upwards of twenty years now the company has been using electric arc welding, 
and it has more recently adopted resistance welding and the oxy-acetylene flame for 
operations for which arc welding is less suitable. An account of what is being done at its 
establishment at Phcenix Wharf, Uxbridge, should, therefore, prove of interest. 

At the outset it may be explained that the output of the works consists not only 
of the steel barrels which the name of the Arm would lead one to expect, but cylindrical 
steel vessels of all descriptions from small drums to large tanks capable of- holding 
3000 gallons or so, and of withstanding heavy internal or external pressures. The 
products include such things as buoys, cylindrical casks and vessels of all sorts and 
sizes, and even, we understand, small steam boilers at times. The material employed 
is steel sheet or plate of varying thickness, and we believe we are right in saying that 
every joint is welded, not a single rivet being used in anything which iA made. 

As acetylene welding does not come within the scope of this Work we may 
dismiss its employment at Uxbridge with a few words, though in so doing we must 
not be understood to infer that the process is regarded by the firm as being of small 
value, for that is far from being the case. Indeed, Mr. T. T. Heatou, the company's- 
managing director, who has certainly had as much experience as, and probably more 
than anyone else in the country in methods of welding other than those practised by 
the smith, regards it as a most useful adjunct to his electric equipment, and ad being 
exceedingly well adapted for certain classes of work — in fact, in its sphere, better than 
any of the electric arc processes. In some cases the particular utility of the acetylene 
flame is due to the fact that its 
temperature is considerably lower than 
that of the electric arc. Hence it can 
be employed with thinner material 
than is practicable, without burning the 
meta], by any but a most experienced 
operator with the carbon arc. 

As far, however, as our own personal 
observation went, on the occasions which 
were courteously afforded us recently of 
visiting the works, the special use to which the acetylene blowpipe, is put is in 
building up work over comparatively wide areas, such as in the joints connecting 
outwardly dished ends to cylinders of large diameter, and made with comparatively 
thick plates. In work such as this it is the firm's custom to build up a layer of 
material round the external joint in ouch a way that the step that is shown in the 
left-hand view of Fig. 12 is entirely obliterated, the joint being made to have the 


"^HC Enoinccr" 


Fig. 12. 



appearance shown in the right-hand view. The width of the welded portion was, in 
the cases which we have in mind, 2 in., or something over. For such operations as 
this, Mr. Heaton has found the acetylene flame to produce better results than the 
electric arc. 

Passing on to steel barrel manufacture, it may be first explained that only the 
carbon arc system is employed, and that it is carried on in exactly the same manner 
as it was when revised by its inventors, Benardos and Olszewski, the " work " forming 
the positive pole, and the carbon the negative pole of a direct current arc. The 
requisite accessories are of the simplest description. There is, first of all, a source of 
electrical energy with a voltage of 90 ; then there are : (a) a resistance of coiled 
wire carried in a wooden frame attached to the wall ; (b) a flexible cable connecting the 
resistance with (c), an electric holder of a very simple type, and (rf) a c&rbon electrode. 
In pre-war times graphite electrodes were employed, but owing to the difficulty in 


•*TmC CNaiHUR** 

Fig. 13.— Carbon Electrode Holder. 

GwAiN Sa 

obtaining them during the war period ordinary carbons are now used, and they 
answer perfectly, though the resistance of the carbon is somewhat higher than that of 
the graphite. The holder consists of a wooden handle with a hole through it from end 
to end for the passage of the flexible conducting cable, and furnished with a heavy 
brass cap or ferrule to which are attached, in addition to the cable, two strips or arms 
of steel, 15 in. long or so. Riveted to the outer end of each of these arms is a stout 
steel plate, an inch or so square and a quarter of an inch thick. Each plate has a 
groove out in it, so that the two plates when pressed together may so clasp the electrode 
that it is held firmly. Pressure is applied by means of a screwed bolt and butterfly 
nut, the former passing through holes in the two clip arms. The apparatus, the 
appearance of which is roughly shown in the accompanying sketch. Fig. 13, is com- 
pleted by a large circular screen some 9 in. in dijimeter, and made of tinned sheet 
steel, which is attached to the wooden handle, and is, of course, for protecting the 
hand and arm of the operator from the effects of the rays emanating from the arc. 
The only other accessory is a hand screen, some 18 in. square, furnished with a handle, 
and having a rectangular opening in its centre measuring about 6 in. by 3 in., which 
is covered by three glasses. That which is held facing the arc is ordinary blue 


glass, and its purpose is only to protect the glasses behind it from splashes of 
metal. Of the other two glasses, one is deep ruby and the other deep blue. We 
gathered that no particular precautions are taken to examine these glasses spectro- 
soopically. They are simply bought, of the darkest shades possible, in the open 
market. The handle is further protected by an auxiliary screen which envelops. and 
protects the left hand of the operator. Each welding space is partitioned off from 
those on either side of it, and from the rest of the shop by screens of sacking. 

We may here say that no ill-effects have been experienced with the eyes, face, 
hands, or arms of any of the operators, when the appliances provided have been 
properly used and reasonable precautions taken. It will be realised^ however, that 
there is some small disadvantage in the use of combined deep ruby and deep blue 
glasses, in that it is quite impossible to see anything through them unless the arc has 
been actually struck. For some little time now the company has been studying the 
question with a view to obtaining glass which, while preventing the passage of harm- 
ful rays, would allow the operator a better view of his work. So far, however, it has, 
we gather, been unable to obtain glass — such, for example, as Crookes' glass — ^in pieces 
large enough to suit the particular requirements. The ruby and blue glasses have 
dene very well for over twenty years, and from our own experience We can say that, 
as soon as the arc is struck it is perfectly easy to watch the welding operations minutely, 
even when standing at some little distance away. 

To the anexperienced onlooker the welding operation might seem to be quite 
sifnple, but, as a matter of fact, it requires no inconsiderable amount of skill. Mr. 
Heaton is of opinion that, taking the sum total of the factors going to make up a 
successful carbon arc weld as one hundred, operating skill would be represented at 
sixty. Yet it is skill which is not, by any means, insurmountably difficult to acquire. 
Some of the operators who have been trained into highly successful workmen had 
had no previous mechanical or scientific experience. We observed two men who were 
doing excellent work whose original occupations had been, we were informed, that of 
milkmen. A man of average intelligence can, we gather, be trained to carry out the 
simpler operations within a period of some three months. <, 

A description of the making of an ordinary steel barrel will give a good idea of the 
general run of this department of the firm's activities. A sheet of steel of the required 
dze and gauge is first of all passed several times through a pair of rolls, so shaped as to 
give the metal the required curvature and form. The edges which are to be welded are 
then accurately marked off and sheared in a machine which was specially designed for 
this purpose. A hole, which is eventually to receive the bung-ring, is then punched. 
The bent sheet is then clipped on to a substantial anvil arm which has the same curva- 
ture as a completed barrel and hence, also, as the rolled sheet. A space of about one- 
eighth of an inch is left between the two sheared edges, which are arranged to come 
immediately over the centre of the anvil arm, which is connected to the positive 
pole of the electric supply. 

The welding of the seam is done in lengths of about 6 in. The procedure is as follows : 
A shearing of steel sheet of the same composition as the sheet being welded, about 
i in. wide, ^ in. thick, and some 6 in. long, is laid flat immediately over the lengtii of 
seam which it has been decided to weld first. Usually it is at one end of the seam, 
though the exact position does not appear to matter* The operator then, with the 
screen in his left hand and the electrode holder in his right, brings th*5 point of the 
carbon electrode into contact with the work, and immediately withdraws it.. The 
result is, of course, the formation of fl,n arc, which, in the operations that wo watched, 
was varied during the process from about 1^ in. to 2 in. in length. This arc is made to 


play on the piece of sheared metal just mentioned, being moved from end to end of it, 
until, in a remarkably short spaoe of time, the whole is a white-hot mass. The operator 
then causes the arc to play on the metal sheet on either side and at the ends of this 
white-hot mass, taking it round and round the latter until the correct welding heat is 
reached, which is judged by the appearance of the metal. The arc is then smartly 
broken by withdrawing the electrode quickly, and the weld is consolidated by a few 
blowd of a wide-faced flat hammer. This cycle of operations is repeated until the whole 
length of the seam has been welded. The seam of a barrel to hold 100 gallons is 
welded in from nine to ten minutes. 

It might be asked why the weld does not extend to the anvil so that the partly 
formed barrel would adhere to it. That it does not do so is due, of course, to the fact 
that, having large mass, the metal of the anvil is not raised to anything near welding 
heat. Moreover, the arc is never allowed to play actually on the anvil itself ; there is 
always some metal in between. 

The art in the welding process is in keeping the temperature just right, neither too 
high nor too low. If too high the metal is burnt, if too low a bad weld will result. 
The temperature is regulated by the current flowing in the arc and by the length of time 
the latter is allowed to play on any particular spot. The current is controlled to some 
extent by the resistance in the external circuit, but not entirely. The operator can, 
however, increase or decrease it at will, over a fairly wide range, by decreasing or 
increasing respectively the length of the arc. He judges by the appearance of the 
heated metal whether the temperature is correct or not, and he acts accordingly. 

It is noteworthy that no flux whatever is used. The company never has used any. 
It maintains that, provided the operation is skilfully performed, there is, for mild steel, 
no necessity to use a flux either to assist in the flow of the metal or to prevent oxida- 
tion. It will be remembered that it is only the strap — ^if suoh a description may be 
applied to the added piece of metal — ^which is actuaUy melted, and, as we saw it, it was 
only just melted and not raised nearly to the burning point. The metal on either side 
of it was not brought to nearly Such a high temperature, as was quite evident from the 
colour. The result, howevOT, is that a perfect joint is produced, which is only very 
slightly thicker than the sheet, the edges of which it unites and which is quite smooth. 

The next process is the welding-in of the bung receiver, which is a ring of mild 
steel, with its hole screwed to receive the bung and with a fairly wide shoulder which, 
when the part with the smaller diameter is inserted in the punched hole, prevents 
the ring from falling through. This shoulder also serves another purpose. There is no 
added metal as in the longitudinal seam of the barrel, but a large proportion of the 
shoulder itself is melted by the arc being made to play on the periphery of the ring. 
At the same time the temperature of the metal sheet of the barrel is raised to such a 
point that fusion takes place between it and the metal of the ring. Hammering is not 
resorted to, and it can well be imagined that considerable skill is required correctly 
to regulate the heat. In some cases the bung ring is only welded on the outside in the 
manner just described, but in others it is welded inside as well. We observed no 
tendency to distortion either in the welding-in of the bung rings or in the formation 
of the longitudinal seams. Moreover, we were informed that no injury was done to the 
screw of the bung ring, though it may be necessary to run a sizing tap through the hole 
if the bung wiU not screw in with sufficient ease. 

The filling in of the ends of the barrel is performed in the following manner : The 
two end pieces are dished so as to form circular trays with rims about 1 in. deep. 
They are of just the correct diameter to be driven into the open ends of the barrels, 
which are left parallel to the axis for a sufficient length for that purpose. The appear- 



ance is then as shown diagrammatically in view " a " in Fig. 14, the thickness of the 
metal being, of course, greatly exaggerated. Two steel rings R^ and R° are then placed 
inside the dished end and outside the end of the barrel respectively. These rings are 
made to fit fairly tightly, but no great care appears to be taken to make them fit 
accurately all the way round. The appearance is then as shown in view '' 6 " in Fig. 
14. The body of the barrel, which is stood on end, is then made the positive pole of an 
electric circuit, and the operator then proceeds to strike an arc, with a carbon electrode 
forming the negative pole of the circuit, on the upper surface X formed by the rings 
R' and R°, the rim of the end piece, and the end of the barrel proper. In this maimer, 
by taking the arc gradually right round the circle, all the four rings are welded together 
as shown in view " c " in Fig. 14, and a stout projecting rim is formed which, from 
the look of it, will stand a good lot of knocking about. Again no flux is used nor 
is hammering required. The time taken to weld round the end of a barrel is about 
1 J minutes per foot. 

Finally, the ends are examined and any rough places removed by filing. As a 
matter of fact, the work is left wonderfully flat and smooth, and only where the metal 
has run over the edges of the inner or outer rings is any attention necessary as a rule. 

End oFBarrei 


**The Engimcer" 



Side or 
Bar re/- 



End oF Barrel 

Side of 


Fig. 14.— Welding-in the End of a Barrel. 

Swain 8e. 

All that now remains to be done is to test the completed barrel for tightness — ^which 
is effect^ by immersing it in water and appljdng air pressure internally — fitting the 
bung and painting. The average time taken to produce a complete barrel, including 
rolling, welding the seam, fitting the bung ring, welding-in the ends, testing, painting, 
etc., is about an hour. 

The procedure with cylindrical vessels with paraUel sides is substantially the same, 
but, of course, the plates are only rolled so as to form a cylinder, and are not bulged. 
If stiffening rings are required they are threaded on and welded in place before the ends 
are welded in. In some cases, too, when barrels are required to be speciaUy strong, 
thick stiffening plates, bent into the form of rings and welded, are put over them where 
their diameter is largest. The rings are bedded to the body of the barrel, and the edges 
finally welded to the latter. The procedure is as follows : A welded ring of eheet 
steel with an internal diameter exactly the same as the largest external diameter of the 
barrel, or cask, which it is desired to strengthen, is slipped over the barrel, ad shown 
diagrammatically in view " a " in Fig. 16. The ring is then connected to the positive 
terminal of a direct current circuit and an arc is struck on one side, say, at " a:," by 
means of a carbon electrode which forms the other pole of the circuit. A compara- 
tively low current is employed, and the arc is not allowed to remain in one place, but 
is quickly moved about over a considerable area until the metal in that area has 
become red hot. The heated portion is then quickly hammered down with a largo 



flat hammer until it comes into close contact with the outside surface of the barrel, a 
suitably shaped anvil arm being, of course, arranged inside the latter. Things are 
then as shown diagrammatically in view '* 6 " in Fig. 15. This process is then repeated 
on the other side, " y," of the ring, so that that side, too, is brought into contact with 
the barrel. These processes are repeated all round the periphery of the barrel, so that 
the bent ring is brought to the same shape as the barrel, as shown in the view " c " in 
Fig. 16. Finally, the edges of the ring are welded aU round to the body of the barrel, 
so that the ring adheres tightly to the latter and provides a most eflEective stiffening. 

It may also be mentioned here that 
when necessary the company cuts metal 
by means of the electric arc. 

The electric energy used in the 
foregoing processes is generated on the 
site. There are two vertical high-speed 
steam engines, one by Belliss and 
Morcom and the other by Browett- 
lindley. They obtain their steam from 
two hand-fired Lancashire boilers, and 
each drives a ^00-kilowatt direct current 
generator. The voltage is 90, and the 
available current is, therefore, about 
4500 amperes. When in full work 
twenty welders are employed, and each 
dynamo supplies current to ten welders. 
The voltage across the welding arc varies, of course, with the resistance in circuit 
and the length of the arc itself, and the variation is usuaUy within the Umits of 50 
to 56 volts. There are no instruments by which the operator can tell what the 
voltage is, or what current is passing. His experience tells him if the conditions are 
as he wants them. If they are not, he can make the necessary correction by altering 
the length of his arc. There are, however, instruments on the switchboard in the 
engine-room by which the operations of each of the workmen can be checked. We 
gather that for ordinary welding work the current employed is somewhere in the 
neighbourhood of 300 amperes. For such a current carbons 15 mm. in diameter and 
12 in. long are employed, and they last for some few hours. 

As to the quality of the work done there can be no two opinions ; it is excellent, 
and it has stood the test of many years' cervice. It is quite evident, therefore, that for 
the particular articles dealt with at Uxbridge the carbon arc process is, in skilful hands, 
and without the use of flux of any kind, perfectly successful and satisfactory. 

We defer a description of the resistance welding system employed in these works 
to a subsequent chapter. 

"The Enoinccr" 

Fig. 15. — Putting on Stiffening Ring. 

Swain 8c 



The business of Pontelec Welding Patents, Limited, dates from early times as elec- 
trical matters go. Mr. A. Jevons, who succeeded Woodhouse and Rawson United 
(Midland Branch), Limited, and who was one of the pioneer contractors in electric 
light and power, established himself in the Minories, Birmingham, in 1888. On the 
introduction of electric welding he was appointed representative for the original 
*' Thomson " patents, and he himself subsequently obtained patents for welding 
the seams of tubes, chain welding plant, and automatic switches for welding machines. 
His office was at Constitution Hill, Birmingham, and the present company, which 
was formed to take over the rights of the Universal Electric Welding Company of 
New York, fitted up a demonstration shop in the immediate neighbourhood. Even- 
tually, owing to the keenness of continental competition, as well as to Mr. Jevons' 
failing health, the company arranged to take over the latter's business, and removed 
its plant to 46 Constitution Hill, at which place, for some years past, demonstrations 
of its various special processes have been given, and also the manufacture of some 
twenty-two different types of machines carried on. 

We have recently had an opportunity of visiting the works and of observing a 
number of machines in operation. Seeing that all the latter embody the Thomson 
method of welding, single-phase alternating current is, of course, alone employed. 
As the public supply available in the neighbourhood of the works is direct current 
it was necessary to employ a means of converting it to alternating current. All elec- 
trical engineers will realise the difficulty of dealing with 'a highly inductive load used 
intermittently and often only at 0.6 power factor, thrown on and off, it may be, several 
times per minute. However, the machine which is employed has, according to the 
testimony of the company^ which has during the past few years purchased several of 
them for use in various places, answered its purpose admirably. It was made by the 
Phoenix Dynamo Manufacturing Company, Limited, of Bradford, which took up this 
particular branch of the subject a number of years ago now, and, as a result, developed 
a special design of inverted rotary converter. 

The actual machine used at the Pontelec Works is of 50 kilovolt-ampere capacity. 
It had, so we gathered, been running for over 8 years, and had been in almost con- 
stant use during that period, frequently being csJled upon to work up to as much as 
70 kilovolt-amperes, and -was still in good working order. The company, in fact, 
was quite enthusiastic concerning its behavioui*, and we certainly saw it undergo some 
fairly severe treatment. 

It would be impossible for us to refer separately to all the many types and designs 
of machines which this firm manufactures,' and we therefore propose to deal with only 
a selection. The machine illustrated in Fig. 16 is typical of the firm's standard type, 
a series of which is built in sizes varying from 6 to 100 kilowatts in capacity. In the 
particular instance shown the working is effected, both as regards the electric and 
mechanical arrangements involved, by means of the hand. In other forms, however, 
the current can be switched on and off by means of a treadle. The whole apparatus 


is wonderfully simple. The electric portion, which, with the exception of the switch 
A and ammeter B at the top, is not visible in the engraving because it is hidden by- 
parts in front of it, consists of a static transformer, the primary windii^ of which 
comprises a number of turns, while the secondary winding is of one turn only. The 
latter m made up of numerous strandB of thin 
sheet copper, which are not only quite flexible, 
but h^ve a negligible ohmic resistance. The 
two terminals of this winding are respectively 
connected to the two jaw carriers C and D, 
whi-jh are mounted in slides formed on the 
front of the machine. These jaw carriers, which 
are, of course, electrically insulated from one 
another, are capable of vertical motion upwards 
and downwards in their slides. The upper jaw 
carrier C is moved up and down by the manipu- 
lation of the hand lever E, which operates 
through the toggle F. The range of movement 
is, therefore, always the some. In order that 
articles of varying thickness may be received 
within the jaws the lower jaw carrier D may be 
moved vertically up and down by means of the 
hand wheel H. The jaw carriers are formed of 
heavy masses of copper, and they are cored out 
so as to permit of water circulation. JawB of 
V, or any other desired shape, can be fitted to 
their faces. 

In operation the article to be welded is 

inserted between the jaws, the height of the 

lower jaw being adjusted so that a very small 

movement of the hand lever will cause the 

work to be gripped. The circuit of the primarj' 

coil is then completed by pulling the switch 

handle rod A^, and current immediately begins 

to flow in the secondary winding, since its 

circuit is completed by the article between its 

terminals. The flow of current is heavy, being 

only limited' by the resistance offered by the 

surfaces in contact, and by the article beiug 

welded. The result is that the latter is verj- 

quickly raised to welding heat, while the jaws, 

because of their mass, and of the fact that they 

have water circulating inside their carriers, 

Flo. 16.— Pontelec EesisUnce Welder. remain comparatively cool, even when large 

numbers of articles are welded consecutively. 

As soon it,s weldiiig heat is reached — and the time it takes to do bo varies, of course, 

with the nature of the article being dealt with, though it is never very long and is 

frequently only a few eecondb — the current is cut off by opening the switch, and at 

the same time additional pressure is put on the work by the hand lever and toggle, 

and the weld is consolidated. We may say that it is impossible to pot excessive 

pressure on the work with the toggle mechanism, since the pressure which can be 



exerted is controlled by the spring G. We have watched the machines of thifi type 
doing a large variety of work, and dealing with articles ranging fr)m thin sheets 
and fairly fine wire up to plates each three-eighths of an inch or so in thickness. 
In some cases the pieces of material to be joined were simply laid one on the other 
without any preparation. No special care, for instance, was taken to remove rust. 

The company has developed and patented several processes for use in connection 
with machines of this tjrpe. One of them, which is Imown as bridge welding — whence 
the name of the firm " Pont-Elec " (trie) — consists of laying small pieces of metal 
either over the joint of a tube or between two separate parts of an article, and placing 
the whole between a pair of electrodes. When the current is allowed to flow and 
welding temperature reached, pressure is applied, the metal is forced together, and 
providing the superimposed pieces are close enough together a continuous joint 
results. In a second process, which is termed multiple point welding, a number of 
dents, or projections, is formed on either one, or both, parts to be united. In this 
case contact only takes place at those points, so that when the current passes heating 
is only set up in their neighbourhood, and when pressure is applied and intimate con- 
tact effected the metal in those parts alone is welded. In some cases as many 
as 20, or more, of these small welds can be made in an article at one operation, and 
the tenacity with which the two portions of the welded article adhere to one another 
can be imagined. For certain purposes machines working on this system can be 
made automatic. The process is, of course, particularly suitable for thin metal work, 
and the electrodeiS, or jaws, may have large flat surfaces. The company draws 
attention to the fact that in this case it is the work or material being dealt with that 
has the points on it, and not the electrodes, so that not only are more uniform results 
obtained than with ordinary spot welding, but the electrodes, being removed from 
the point of maximum temperature, are little affected, whereas with single-pointed 
electrodes the copper points which, by reason of the high temperatures and the 
hammer action to which they are subjected during the welding operation, are apt to 
become flattened, which necessitates their being frequently attended to and reshaped. 

A third process is particularly intended for use in connection with heavier work 
such as the welding together of plates of half an inch or more in thickness. In such 
operations it is an exceedingly difiScult matter to maintain the points of the electrodes 
in good condition, even though the electrodes themselves are as efficiently water-cooled 
as it is possible to make them. There is, moreover, another trouble in that, owing to 
the plastic character of the metal being welded when heated, indentations on both 
faces of the plate may be caused by the heavy pressures which are necessary to ensure 
good welds with thick plates. Disc welding, as it is termed, has been devised to over- 
come these troubles. It consists in placing small discs of the same metal as that which 
is being operated on, on one or both sides of, and in some cases between, the parts 
being welded, the first two discs being arranged at the points of contact between the 
electrodes and the work. On the passage of the current, the parts heated are in this 
way restricted to the area of the discs or something a little greater. Welding heat 
is by this means obtained very quickly, and when pressure is applied the discs are 
forced into the work. By suitably arranging the diameter and thickness of the discs 
a flush surface results. If desired, too, bosses may be put on, and such bosses may be 
shaped like rivet headd so as to give the finished article the appearance of having been 
riveted. The company is careful to explain, however, that the strength would be 
greater than that possessed by riveted work since the area of the weld is always some- 
what greater than that of the superimposed disc. A particular advantage claimed 
for this method is that greatly varying thicknesses of metal can be welded together 


without any chance of burning the thinner sheet. Thus by it, it is perfectly feasible 
to weld a thin sheet of, Bay, 26 or 30 gauge on to a plate 1 in. or 1^ in. in thickness. 

Still another process which can be practised with this machine is that known as 
ridge welding. In it ridges are formed during rolling on the surfaces of the metal, so 
that when the two articles to be welded are brought together the area in contact is 
much reduced as compared with what it would be were the Rurfaces plain, and welding 
only takes place along the ridges. This process, it is pointed out, is especially useful 
in constructional work, but has not received in this country the attention which has 
been accorded to it both on the Continent and in America, 

We understand that a Ucence is required to use the special processes of this com- 

Fio. 17.— Machine for Brazing Collars on Tabea. 

pany, even though they are carried out on machines of another make, since the pro- 
cesses themselves are subjects of patents. 

Passing on now to some of the other machines made by this firm, attention may 
be directed to Fig, 17, which shows a machine recently designed for brazing steel 
collars on to tubes. In this case the electrodes have horizontal instead of vertical 
motions, the pipe being dealt with being held vertically. The pipes seen in the en- 
graving are for the water circulation, and give the machine the appearance of being 
rather complicated, which, as a fact, it is not. 

It will be observed the welding circuit is divided and forms three pairs of jaws, or 
tools as the makers call them, the upper and lower pairs fitting the tube, whilst the 
middle pair makes contact on opposite sides of the coUar to be brazed on. In opera- 
tion, the handle an the right-hand side is pressed back and all three tools on that aide 
are carried to the righti the left-hand tools being fixed. The collar is first placed in the 


middle tool, resting upon. Bupports provided on the Inside of the tools. A ring of 
br&zing wire is then placed on the collar and the tube passed through both from the top, 
an adjustable support at the bottom — not seen in the engraving—taking the weight 
of the tube and at the same time fixing the position of the collar in relation to the ends 
of tube. The handle being then pulled forward, all three pairs of tools are brought 
into contact with their respective parts, the secondary circuits being so arranged 
that one is completed through the tube, which becomes heated midway between the 
upper and lower contacts, whilst the other circuit is completed through the collar, 
which becomes heated at the same time, there being only one primary circuit con- 
trolled by a foot-switch so that the operator has both hands at liberty. A little borax 
is applied and the current kept on until the brass wire flows into the joint between the 
tube and the collar. Owing to the heat being internally generated, the outside of the 
collar remains clean and nothing is needed, so we understand, but a rub over with a 
wire brush, the whole operation only occupying about one minute. The tubes are about 
one and five-sixteenths inches external diameter and one-quarter of an inch thick, 
the collars being about one and one-quarter inches long, and three-eighths of an inch 

Fig. 18. — Speclacle-frame Soldering Macliiae. 

thick. The regulator provides six variations of power, with a maximum of 10 kilovolt 

Fig. 18 is interesting as shewing what is thought to be the smallest resistance 
heating plant ever made commercially. It is not, strictly speaking, perhaps, a welding 
machine any more than is the machine just described, but they both approach, we 
think, sufficiently nearly thereto to warrant mention of them being made in this 
Work. The whole apparatus is contained in an oak case which measures 10 in. each 
way, and it weighs only about 28 lb. It was designed and built for the Army Spectacle 
Department, and is used, so we are informed, for the hard soldering of spectacle frames. 

The machine on the left is a 4 horse-power motor rewound, and slip rings added. 
The transformer is enclosed in an oak case, which has a regulator attached to its side. 
The small switch shown in between the two is intended to be foot-operated, and is 
placed on the floor, the converter being arranged in any convenient position. The 
small machine shown in front of the oak case consists of two end frames supporting 
a pair of parallel bars carrying a pair of copper blocks, both insulated and connected 
respectively to the two 'ends of the secondary winding, but capable of being moved 
on the bars and fixed in position by a small thumbscrew. Each copper block carries 


a tool formed to hold the different parts, which are pressed together by a light apri ng. 
On the foot>Bnit«h being closed the points of contact become heated, and hard solder, 
in the form of a fine wire, is touched upon the heated parts, inth the result that it is 
melted and, flowing in, forms a clean and neat joint. Several hundreds of spectacle 
frames are, we are informed, completed by one operator per hour, whilst the heating 
is eo local that ndjoining parts are not softened, and the articles are easily handled. 
This machine has a maximum capacity of about 2 kilovolt-amperes. It is use<l on 
both Kteel and gold-SUed fittings of spectacle frames. 

Fig. 19.— Typical Pontelec Spot Welder. Fio. 20.— Spot Welder for Bench Work. 

Spot welding machines of various types are made by the company. A typical 
example of one of them is shown in Fig, 19. It is of a light type, and is operated by 
pedal. It is intended speciaUy for welding hollow-ware previous to enamelling, and 
is mad© in sizes ranging from 5 to 20 kilowatts in capacity. As will be observed, the 
lower electrode or anvil is capable of adjustment vertically by moving it in its slide, 
where it can be clamped in any desired position by the nuts and bolts which may be 
seen in the engraving. The upper electrode can also be adjusted by the movement 
of a nut up and down the vertical screwed spindle seen at the rear of "the pivoted arm. 
The distance apart of the two electrode points, and hence the amount of pressure 
applied during welding, can in this manner he arranged. 


A much lighter type of spot welding machine for bench use is shown in Fig. 20, 
It also is pedal worked, and it will be observed that the lower portion of the link con- 
necting the pedal with the upper electrode is slotted, so that the latter can be raised 
to admit larger or smaller work. Then, too, the amount of travel of the pedal, and 
hence of the electrodes, can be regulated to suit the altered circumstances by means 
of the two sets of screws seen at the bottom of the engraving. The spring connecting 
the frame with the pedal operating link raises the latter, and hence the upper electrode 
when presBure is t>aken ofi the pedal. It will be observed that there are two series of 
terminals on the box at the side of 
the framo. They are for employing 
different tappings on the primary of 
the transformer inside the machine, 
so as to obtain different voltages in 
the secondary to suit differing classes 
of work. This, we may mention, is 
an attribute possessed by the other 
machines made by the company. 

While on the subject of spot 
welding machines, we may say that 
on the occasion of our visit to the 
Pontelec Company's works we were 
shown the drawings of a very large 
machine of that type — probably, it 
-was thought, the largest which had 
been undertaken in this country — 
which was then under constraction 
for an enterprising British firm. It 
was to have copper arms 7 in. in 
diameter, projecting 5 ft. clear from 
the body of the machine, and its 
weight, when completed, was to be 
something over 5 tons. An interesting 
insight into the range of machines 

built by the firm may be obtained ^lo. 21.— Pontelec Butt Welder. 

when comparison is made between 
that weight and that of the spectacle-frame machine weighing 28 lb. 

Fig. 21 shows one of a series of butt weldii^ machines, which are made 
in sizes varying from 6 to 60 kilowatts capacity. The actual machine illustrated 
had been specially designed for welding motor-cycle rims, which it does, we gather, 
at the rate of 150 per hour, with a consumption of 12 kilovolt-amperes of energy. In 
operation the two ends of the rim which are to be welded are gripped between 
two specially formed jaws, which can be moved backwards and forwards 
horizontally, so that pressure can he applied to the weld by means of a hand lever 
and toggle mechanism, observable on the right of the picture. The electrodes have, 
of course, water circulation. We have seen this machine in operation doing excellent 
work. It is fitted with an oil break automatic out-ofi switch. 

A machine of quite a different type, which nevertheless shares many features in 
common with those already referred to, is shown in Fig. 22. It is a machine for welding 
the seams of oyUnders, and its operation will be reaiUly understood. The cyfinder is 
fixed in a frame which can be traversed backwards and forwardn by means of the 


hand lever. The seam restB on a horizontal anvil which forme one of the electrodes, 
and has its ends carried by uprightg which form part of the cylinder-holding 
frame. The upper electrode is formed by a copper roller which is revolved by 
power by means of a chain which in sufficiently slack to permit of the small 
amount of movement necessary to bring the wheel into contact with the work. Both 

anvil and wheel are water cooled. 

When the cylinder is in position, the 

bed and frame carrying it are moved 

by the hand lever, so that one end 

of the Beam may come directly under 

the roller. The' latter is then depressed 

until it just touches the work. The 

pedal bwitch is then operated so that 

current flows, while simultaneously 

the pressure between the roller and 

cylinder is increased, with the result 

that the latter is gripped between 

the anvil and the roller, and, by 

reason of the rotation of the latter, 

ia caused with its bed and frame to 

move horizontally. The result is that 

the wheel follows along and welds the 

seam as the cylinder is gradually 

traversed from one side to the other 

of the machine. As in the machine 

shown in Fig, 16, there is a spring 

which prevents excessive pressure 

beii^ put on the work. In giving 

the foregoing descriptions of eelec- 

tions of the machines made by this 

firm, it is not pretended that all the 

different types of machine which it 

makes have been mentioned ; but a 

fairly rejM^sciitative reference has 

been made. The company, it may be 

Fio. 32.— Typical Potitelec Seam Welder. mentioned, does not issue any 

catalogue. It has found such a course 

impossible because there is such a great variation in the voltages, periodicities and 

systems of distribution, that hardly any two cases are ahke in every respect. In 

some instances, for example, only polyphase current may be available, and it is not by 

any means always permissible to put a welding load on one phase only. Then, again, 

continuous current only may be available. It prefers, therefore, to treat each separate 

case on its own merits, and is prepared, when supplied with full information, to give 

its considered view of the matter ; and it lays stress on the fact that it makes a point 

of never booking an order unless it is sure that a commercially satisfactory result 

can be obtained. 



As an example of coated metal electrode processes we have chosen that which is 
known as the Quasi-arc process, and we discuss it in the following chapter : — 

The Quasi-arc — or the " Quazarc," to lise the trade mark name that has been 
adopted — process of electric welding is the invention of Mr. Arthur Percy Strohmenger, 
of the Quasi-Arc Company, Limited, of 3 Laurence Pountney Hill, E.G. 4, who holds 
several patents in connection with it. It is a fusion process, and the electrodes used 
are coated in a special manner, both they and the method of their application having 
been patented. The electrodes can only be obtained direct from the company, and 
with each parcel is granted to the purchaser a direct licence for the use of the process 
in conjunction with the electrodes supplied in that parcel, and for such use no royalty 
is charged. The firm has extensive works in the East of London, and at them we have 
had an opportunity of observing, not only the manufacture of the electrodes on a large 
scale, but the actual process of welding in operation. Mr. Strohmenger, who, we be- 
lieve, was bom in South Africa, has had, at any rate, a wide experience with the 
asbestos deposits of that country. Our readers will remember much of the asbestos 
found there is of a blue colour, and is largely composed of a ferrous silicate. Its 
chemical composition is such that Mr. Strohmenger was led to the belief that it would 
answer perfectly as a flux in electric welding. His first idea was to coat a metal rod 
with the asbestos and to lay it along the joint or seam which it was intended to weld. 
An arc was then struck by means of a carbon or other electrode, and both the coating 
and metal rod were gradually melted, together with the edges to be welded, so that a 
welded joint protected by a covering of flux or slag was obtained. We believe that 
the character of the weld was all that could be desired, but the method was clumsy and 
did not, moreover, permit of overhead working. So Mr. Strohmenger decided to use 
the coated rod itself as the electrode, and he has since made certain improvements 
in the manufacture of the electrodes themselves. Perhaps the most important of these 
improvements is the introduction between th^ coating and the rod of a fine aluminium 
wire which represents in bulk about two per cent, of that of the electrode metal itself. 
It is claimed that the effect of adding the aluminium is that a strong reducing action 
is brought about, the metal having a strong affinity for oxygen at welding temperature. 
The asbestos itself acts as a reducing agent and forms a slag which covers the weld 
and prevents oxidation. The alumina passes away with the slag and the coating 
formed on the weld is easily chipped off by hammering, and removed with a stiff brush 
when the metal has cooled. The temperature at which the coating melts can be con- 
trolled to a certain extent by the addition of such compounds as aluminium silicate 
or sodium silicate. We may say here that the coating is applied in the form of yarn, 
and is so regularly put on that the surface of the rods is quite smooth to the touch. 

Under the Quasi-arc System either single phase alternating or direct current can 
be employed. When we saw it in operation direct current was used. The supply 
voltage was 1 10, and beyond the special holder for the electrode the only other acces- 
sories were a resistance and an ammeter, and, of course, the protecting screen. The 

E.W. D 


negative pole of the current supply ie connected to the work or to the metal plate or 
bench on which the work to be welded rests, though in the latter case it is, of course, 
neoeesary to make certain that there is good electrical connection between the work 
and the plate. The positive terminal of the electric supply is connected to one ter- 
minal of the resistance, the other terminal being connected by means of a flexible 
cable with the electrode holder. One end of the electrode is left bare of coating and 
it is, of course, that end which is imerted in the holder. In working, the electrode 
IB first of all held in a nearly vertical position — supposing the work to be horizontal — 
and when the tip of the electrode ia brought into contact with the work an arc is 
formed. The electrode, still kept in contact with the work, is then dropped to an angle, 
and the are is destroyed owing to the covering passing into an igneous state, and, 
as a secondary conductor, maintaining electrical connection between the work and 
the metallic core of the electrode. The action once started, the electrode melts at 
an uniform rate, so long ae it remains in contact with the work, and leaves a seam of 
metal, which, if the operation has been properly performed, is perfectly fused into 
the work. The covering material of the 
electrode acts as a slag and spreads over 
the surface of the weld as it is formed, 
and it is claimed that the fused metal is 
thereby protected from all risk of oxida- 
tion. It is recommended that the operator 
should commence at the furthest point of 
the joint away from himself, and bring 
the electrode towards himself with a 
V ^ slightly to-and-fro movement so as to 

^-Enlarged diaarsm of Weld siH!yi,«gm(,t,o'ia,<'e- Spread thc heat and the deposited metal 
THt e™.,.." «""«■ ^^_ ^ equally to both sides of the jomt. This 

Fio. S3. zigzag movement, and the appearance 

of the weld formed by means of it, are 
shown in the accompanying engraving. Fig. 23. The length of the arc varies, we 
gather, between ^ in. and J in., depending on the size of the electrode used. 

It is necessary, of course, that care should be taken to feed down the electrode to 
the work at the same rat© as the former melts away, and the operator is warned 
not to draw the electrode away from the work so as to form a continuous arc, as by 
doing so the quality of the metal laid on will be impaired, and the work, if thin, may 
be burnt. The aim, it is explained, should be to keep the electrode just in the molten 
elag, and this may be done by the " feel " of the covering just rubbing on the work. 
Care must also be taken to see that the " work " is fused or melted where metal 
is being depouted. It is quite possible, as we ourselves observed, to distinguish 
between molten slag and molten metal when using the Special screens employed by 
the company, the metal being dull red in colour, and the slag very bright rod. 

In discussing the sphere of its applications the company states that the Quasi-aro 
process can be used for a variety of purposes, such as constructional eteel work, in- 
cluding shipbuilding, in the manufacture of pressure tanks, air receivers in boiler 
work, including the reinforcing of worn plates and general repairs, and in the making 
up of crank-shaft journals, worn key beds, repairs to hydraulic rams. It claims, too, 
that the ease with which seams of any length and thickness can be welded renders 
it most useful in industries in which hitherto riveting and caulking has been solely 
employed — see Fig. 24 — and that on plate work it shows a high percentage of saving 
in cost as compared with oxy-acetylene welding, being more than twice as rapid. 

"The Encimeer" 8<vmm Sc 


and the resulting joint much stronger, while there are no difficulties owing to distor- 
tion of the work. On this point, which is due to the heat being highly localised, great 
stress is laid. With regard to the foregoing it may be said that practically identical 
claims are made by other arc welding specialists, who would hence dispute that they 
are peculiar to the Quasi-arc process. They may therefore be taken as being indica- 
tive of what can be done with metallic arc welding in general. 

The Quasi-arc electrodes ad supplied are in lengths of 18 in., and they are of various 
diameters, according to the size and nature of the work for which they are required. 
For the general welding of mild steel or iron, constructional work, tanks, wheel tyres, 
filling steel castings, etc., mild steel 

electrodes are supplied and in the ^vifeid 

following sizes : S.W.G. Nos. 14, 12, 
10, 8, 6, and 4. For reinforcing worn 
parts of machinery, building up the 

teeth of gear wheels, reinforcing tram- Fig. 24.— Welded Riveted Joint, 

way rail treads, etc., there are carbon 

steel electrodes. As being suitable for reinforcing or building up manganese steel 
crusher jaws, dredger bucket lips, tramway points and crossings, manganese steel 
electrodes are recommended. They are made ^^ in. in diameter. Finally there 
are special electrodes for boiler plate reinforcement and any overhead or vertical 

With regard to the sizes of electrode required for different sizes of work in sheet 
and plate welding the company publishes a table which we reproduce on page 37. 

It will be noticed that in thick work no attempt is made to do the whole depth of 
the weld at one operation. The process is, in fact, essentially a building-up process. 
Two, three, four or even more " runs " may be required, a proportion of the depth 
being filled in at each " run." For the bottom portion a comparatively small electrode 
is used, so that its point may be taken down to near the bottom of the seam. It 
need hardly be said that the slag from a preceding '^ run " must be chipped or brushed 
away before a further " run " is commenced, and that care must be taken to fuse the 
surface of the preceding layer of metal as well as the sides of the joint whilst depositing 
new metal. 

As regards special precautions to be taken whilst welding by this process, the 
company recommends that the electrode holder should be gripped firmly, but the 
arm left free ; a light touch and a lateral movement from the wrist give the best 
results. Sufficient current should be used to admit of the metal, which is being 
d^eposited, running freely, with no tendency to pile up. If too much current be used 
the metal will run fiercely and be uncontrollable. The point of the electrode should 
be kept weU down to the work, just in contact, in fact, with the molten slag, and at 
such an angle as will keep the molten metal and the slag flowing in front of the electrode, 
that is to say, away from the operator, who, it will be remembered, gradually draws 
the electrode towards himself. Care should be taken not to allow the slag to get 
behind the electrode, that id between it and the operator. We have already explained 
that slag can readily be distinguished from metal by reason of the (Ufference in 

The following table gives the limits of current admissible with various sizes of 
Quasi-arc electrodes : — 



Current limits. 

Size of electrode. 


No. 14S.W.G. . 

.. 15 to 40 

» 12 

.. 35 „ 90 

„ 10 

.. 75 „ 110 

„ 8 

.. 90 „ 140 

» 6 

.. 110 „ 175 

„ 4 

.. 140 ., 200 


.. 150 „ 220 


.. 250 „ 300 

The following remarks which the company makes regarding special applications 
of the process are of interest : For reinforcing worn or weak work and for the repair 
of faulty castings, the size of the electrode wiU depend upon the amount of work to be 
deposited, but, in the case of work liable to distortion, overheating by the use of 
unduly large electrodes must be avoided. In other words, it is a mistake to attempt 
to lay on too much metal at one operation. For reinforcing work, such as the repair 
of a worn crank shaft, the new metal should be deposited in strips parallel to the shaft, 
and each strip should be cleaned of slag, etc., before the next is deposited at the side 
of it. Then again, before commencing to fill up blow-holes the work should be pre- 
pared by opening it out sufficiently with a chisel, or by fusing away the overhanging 
portions with an electrode — using a higher current than for welding — so as to enable 
the bottom of the hole to be reached by the electrode when welding. 

Overhead welding, for which, as mentioned above, special electrodes are pro- 
vided, takes longer than a seam of similar length on the flat. Good, sound homogeneous 
metal can, it is said, be fused into the joints by holding the electrodes nearly vertical 
and moving slightly from side to side, keeping the point of the electrode close to the 
work. The weld should then be chipped over until an even surface is left, and then 
other " runs " put on until the required thickness is obtained. For vertical welding, 
the ordinary mild steel electrode may be used up to No. 8 size, the best angle to hold 
the electrode being just below the horizontal. Most operators, the firm states, obtain 
the best results by commencing at the bottom of the joint and working upwards, 
although the process should be reversed when welding in an angle fillet. When suffi- 
cient metal has been put on, if the electrode is worked downwards from the top with 
a sHghtly higher current, it will have a smoothing effect on the metal previously de- 
posited. We may say that, as far as we are aware, hammering is very rarely, if ever, 
used with this process. The seams which we saw, whether they were horizontal, 
vertical, or overhead, had a rippled surface — caused, doubtless, by the zigzag move- 
ment given to the electrode — which stood up shghtly above the surface of the work. 
When the slag was chipped off ?ifter welding, the deposited metal was perfectly bright 
and not in the least discoloured. 

For welding cast iron it is best, the company states, to use a small diameter elec- 
trode and a low current, so as to retain as large a percentage of the silicon as possible 
in the neighbourhood of the weld and thus keep the metal soft. Good grey iron, 
high in siUcon and low in phosphorus, welds the best. In welding heavy cast iron — 
say, above | in. thickness — ^it is best to vee out the crack, weld in at the bottom 
with a small electrode, and then build up with successive layers of metal. In welding 
light castings of box section, it is advisable to heat the casting to a black heat, to 
relieve local strains, and then to weld over the crack without vee-ing. This operation 
will result in the welding going quite half-way through. If the underside of the crack 
is accessible it is advisable to repeat the operation from that side. 

Though not strictly coming within the scope of this Work we may briefly 



refer, here, to cutting metals by means of the Quasi-arc process. For this purpose 
the company recommends the use of a mild steel electrode of No. 8 gauge. Just before 
commencing operations the electrode is dipped in water, then, the resistance having 
been set to give a current of about 200 amperes, the point of the electrode is applied 
to the plate to be cut. The molten metal is allowed to drop away, and the cut formed 
is followed up by feeding in the electrode and moving the point of the latter quickly 

Sizes op Quasi-arc Electrodbs and Current Required for Sheet and Plate Welding 

Thickness of 




Preparatiou of joint. 

Close joint 





i in. plate 

I and I in. 

16 s.w.g. 

/ij^ in. plate I j'^^ in. open 

I in. plate I Veed half-way through and -^^ in 

yV hi- and f in. ^ Veed right through to 70 deg. and x^tf in. open 

Two runs are necessary — 
First run 
Veed right through to 60 deg. and J in. open . . 
Two runs are necessary — 
First run 
Second run 
Veed right through to 60 deg. and J in. open . . 
Three runs are necessary — 
First run 

Second run 

Third run 
Over I in. Prepare thus : — 


Up to f in. carry the 60 deg. angle, then parallel 

upwards through the thickness of the plate. 

After welding as for | in. fill up with 
For heavy plates, say, l^in. thick, it may be 

sometimes advisable to vee half-way 

through from both sides 

Size of 

No. 14 










in amp. 

20 to 

30 „ 






















150 „ 175 

up and down through the thickness of the plate. During the process the electrode 
is again dipped once or twice in water. The point of the electrode should always be 
kept as close as possible to the metal to be cut, but not so close as to get a dead short 
circuit. If that occurs the electrode will stick and get red hot. 

As regards the ease with which the operator can learn to employ this process we 
are enabled to give the following testimony of an independent witness who is entirely 
unconnected with the company. He says : " Electric welding by this method is not 
difficult to learn, and with good instruction, skill in it can be acquired, and fairly good 
welds produced, after one or two weeks of practice, according to the abilitv of the 


learner. ,The work is Buitable for female and other unskilled labour. The welder 
may either stand or sit to the work. If the weld is of long duration it is advisable 
to provide seats, particularly for female operators, so as to make them as comfortable 
as possible. Neither the quality of the welds nor the speed at which they are effected 
suffers from this attention." Here again, we would say, that advocates of other 
covered electrodes make practically identical claims for their products. 




To say that the British Insulated and Helsby Cables, Limited, of Prescot, Lanes., 
makes an infinity of machines for electric welding on the redstance system would, 
of course, be an exaggeration, yet one is almost tempted to make such an assertion 
when contemplating the multiplicity of designs which it has evolved for effecting 
numerous operations. 

We shall not attempt, therefore, even to give a full list of, much less to describe 
all the machines which it is prepared to supply, though such a description would be 
full of interest, but shaU content ourselves with referring to what may be termed a 
fairly representative selection. We may premise what we have to say by explaining 
that the electric arrangements involved in all the machines are f undamentaUy the same 
as those employed in all systems of resistance welding. That is to say, each machine 
embodies a transformer wound so as to perform the particular work desired under the 
conditions of voltage and periodicity of the supply current available. The secondary 
coil of the transformer consists of a single convolution, having a large cross section of 
copper, which terminates externally in two electrodes, which are of various forms. 
The pieces to be welded are brought between these electrodes, thus completing the 
electric circuit of the secondary coil, so that when the primary circuit is closed a heavy 
current flows in the former. As the resistance to the flow of that current is practicaUy 
all centred in the surfaces in contact — since the ohmic resistance of the secondary 
winding is compai'atively negligible — great heat is developed between the electrodes, 
and the material between them is quickly brought up to welding temperature. There 
is in each machine provision for regulating the pressure at the points of contact, and 
also for cutting off the current at the right moment. The latter operation is, in some 
cased, performed automatically, though in others it is effected by a separate motion. 
Generally speaking, too, there are means of altering the current in the secondary by 
means of tappings on the primary winding. Such in brief outline are the general 
principles underlying the working of the machines, which we can now proceed to refer 
to in greater detail. 

In the first instance attention may be drawn to butt welding machined, of which 
the company makes various types to suit different purposes, the princip€il being : 
(1) Welders for wire with automatic upsetting gear for uniform section iron, steel 
or non-ferrous metals ; (2) welders for manufacturing purposes with hand or automatic 
upsetting gear for irregular section, iron, steel, or non-ferrous metals ; and (3) chain 

" Wire welders " is the term which the firm itself applies to the machines in cate- 
gory No. 1, and some of the machines made are capable of dealing with wire with as 
small a diameter as 0.024 in. Other and larger machines of the same type are, how- 
ever, able to weld material up to 1 square inch in cross section. Such machines are all 
fitted with automatic upsetting and current control gear. With them the parts to be 
welded are clamped between the jaws of the welder and the ends held together under 


spring preBsure. When the current ia switched on beating occurs at the junction to 
the two pieces, and when the heat is high enough to penuit of it, the ends are forced 
together into still more intimate contact by the action of the spring and the current 
is automatically switched off. Wires joined in this manner are subsequently redrawn 
BO as to do away with the thickened portion where the upsetting has taken place. 

The machines in the second cat^ory are used for such manufacturing purposes, 
amongst a host of others, as welding pipes, refrigerator coils, milk-can rings, peram- 
bulator rims, printers' chases, fittings to casement frames, carriage and coach work 
parts, trellis work, coupling links, brake rigging, travelling-bag frames, low grade steel 
shanks on to high grade steel tools, drills, taps, etc., lubricators, valves. t\-res, etc., 

FlO. 25.— B.I.W. Butt Welder, No. B7. 

and for the building up of drop-forged parts into complete forgings. They work very 
much as do the co-called wire welders, saving that in the majority of cases the switch 
and upsetting gear are hand controlled instead of being automatic. 

The chain wekiers really form a class by themselves, though the general principles 
involved are very much the same as those of the other machines, as we shall show when 
we come to describe them. 

Two examples of the firm's butt welding machines are shown in Figs. 25 and 27. 
The first-named is what is known as welder No. 57, and it has been designed to deal 
with welds up to 5 square inches in cross section, and to be suitable for welding bars, 
tyres, etc. It will be observed that the jaws open and shut horizontally, and that the 
pressure is applied hydraulically, the electric current being turned on and off by means 
of the foot pedal. There is, of course, a water circulation for cooling. The transformer 


is contained in the cast iron base, and different tappings on the primary can he em- 
ployed by meanB of the switch, shown at the left-band side, for work of different 
chsractere. Eye-bolts are provided at the top of the casing, so that the whole plant 
can be slung and carried to the point where it is desired to use it. 

The machine shown in Fig. 27 differs from the foregoing in several respects. It is 
smaller, for one thing, being only intended to deal with welds up to 2 square inches 
in area, and the pressure is applied by the spoked hand wheels acting on a screw and a 
serieB of levers. The current is turned on and off by means of the hand switch on the 
left-hand side. The articles to be welded are gripped in the jawe by the two hand 
vheels, and the jaw carriers, which are of stout proportions, are all of them provided 
with water circulation. The machine is particularly intended for welding drills. 

The manufacture of chains from coils of wire is carried out by two machines, the 
first of which bends the links and threads them into a chain, while the second forms 

Fio. 26.— B.I.W. Cham-Liuk-Fonniiig Machine. 

the welds. Although the first machine does not correctly come within the scope of 
this Work it is necessary to describe it, since the complete process cannot properly 
be understood without it. Three sizes of machine are made. They deal respectively 
with wires having diameters of from j^in. to ^in., ^^in. to^in., and ygin. to-j^in. 
The smallest machine turns out from fifty to sixty links per minute, and takes from 
1 to 2^ horse-power to drive it ; the middle size machine can make from forty to fifty 
links per minute, and requires from 2 to 4 horse-power ; while the output of the largest 
machine is from twenty to thirty links per minute, and a horse-power of from 3 to 7 
is needed to drive it. One of the machines is shown in Fig. 26. The minimum propor- 
tions of the links made on these machines are — length 5 diameters of wire, and width 
3J diameters of wire. The machines are belt-driven and provided with fast and loose 
pulleys. The control is by the hand lever seen at the right-hand or driving end. 
There is also a foot-operated lever which is attached by chain to the hand lever, and 
which, when depressed, throws the belt on to the loose pulley and brings into opera- 
tion a brake which acts on the Sy-wheel, thus enabhng the machine to be stopped 


almost instantly in case of emergency. Power is transmitted to a back shaft through 
a pair of double helical wheels. Mounted on the back shaft are three cams, the first 
of which is a triple-path cam, which operates all the tools forming the links. The 
centre path pushes forward the main slide, which carries a parting-off tool and two 
bending tools, which form the link into a U shape from the back et the mandril. The 
two side paths operate levers carrying the tucking tools, which close the Unk at the 
front of the mandril. The second cam on the back shaft operates a gripping lever, 
which holds the wire while it is being parted off. The length of the path of this cam can 
be varied. The adjustment for gripping different diameters of wire is made by a screw 
at the gripping end of the lever. The third cam operates the feed. It is made adjust- 
able, so that any length of feed can be obtained to suit any link within the range of tiie 
machine. From it the feed slide is operated through a bell crank lever, the feeding 
end of which is fitted with a device which grips the wire on the feed stroke of the slide 
and automatically releases it when the end of the stroke is reached. This gripping 
device is fitted with a smaU cam which enables it to be thrown out of action at any time, 
but the main use of which is to allow the machine to be started up and the motions 
put into operation before actually starting to make chains. The cam is fitted with a 
lever for finger and thumb operation, and should be put in action when the slide is on 
the back stroke. 

Running from the front to the back of the machine on the right-hand side is a shaft 
driven from the back shaft through a pair of bevel gears. Mounted on it are two cams 
and a spur pinion. The first cam operates levers which raise and lower the top mandril, 
an operation which allows the formed link to be taken out by fingers. The second cam 
is responsible for the in-and-out motion of the lacing spindle. The pinion drives a 
spur wheel, in the face of which is sunk a cam path. From this path motion is given to 
a quadrant which drives the lacing spindle in a semi-rotary manner. The semi- 
rotary movement is done in two stages, each of 90 deg. The stroke of the lacing 
spindle is adjustable. 

The mandril is in two parts, upper and lower. The latter is stationary, and only 
acts as a stop and stiffening block for the upper mandril, round the bottom end of which 
the link is bent. A groove is formed down the front of the lever for convenience in 
lacing. An adjustable cutting die is fitted so that the wire may be parted off clean. 
On the extreme left of the machine is a straightening device. 

In operation the first 2 ft. or 3 ft. of the coil of wire are straightened by hand, and 
the end is threaded through the straightening rollers — ^which are slacked off for the 
purpose — through the gripping device on the feed slide, and through the cutting die 
until it — ^the end — ^projects through about J in. on the mandril side. The straightening 
device is then adjusted, after which the fly-wheel is turned round until the shearing 
tool cuts off the projecting | in. of wire. After adjusting the feed grip, main grip, 
length of feed and stroke of lacing spindle to suit^the wire and link, the machine is 
ready for work. The belt is first thrown on to the fast pulley. Then, when the feed 
slide is on the back stroke, the small cam is slipped over and the feed grip thus put 
into action. The machine then works automatically. 

The cycle of operations is as follows : On the forward stroke of the feed slide the 
wire is pulled through the straightening rollers and the end is fed behind the mandril. 
As the slide reaches the end of the feed stroke, the main grip comes down and holds 
the wire on the anvil. Meanwhile the back sUde is moving forward, and as soon as the 
wire is gripped the shearing tool cuts off the length of wire necessary to form the Unk. 
This piece of wire is caught by the back-bending tools and bent into a U shape round 
the mandril. Next, the tucking tools come into operation and bend the ends of the U 


towards each other, thus completing the link. The lacing spindle then moves forward, 
and the* fingers close on the completed link. As soon as this happens the top mandril 
rises to allow the link to be withdrawn. When it is clear the mandril drops down and 
the wire is fed behind for the next link. Simultaneously, the lacing spindle makes a 
quarter of a turn, thus changing the position of the first link from the horizontal to the 
vertical plane, and then moves forward again to enable the fingers to place the link in 
the groove in the mandril. When the second link is formed it is threaded into the first 
link. The fingers then retire and are again turned through an angle of 90 deg. to the 
horizontal plane ready to take out the second link. The motions of the lacing spindle 
are so arranged that the joint on every second link is on the same side, so that the chain 
when being welded has only to pass through the welder twice. 

We can now pass on to the welding machines, one of which is shown in the engrav- 
ing Fig. 28. Three sizes of machine corresponding with the three sizes of bending 
machines are made, and there is a modified form of each of the two larger machines, 
but the modifications are only in detail. The three machines require from one-half 
to 1 horse-power ; from 1 to 2 horse-power ; and from 2 to 3^ horse-power respectively. 
The first can turn out from 15 to 20 links per minute, the medium-sized machine — 
and our illustration shows one of them — ^from 10 to 18 links per minute, and the 
largest machine from 5 to 8 links per minute. 

Each machine consists of a cast iron stand on which i§ mounted a single-phase 
transformer, and all the necessary working parts. Power is supplied by a belt drive 
on to a fly-wheel pulley. Spur teeth, cut on the inner side of the fly-wheel rim, drive 
a pinion and half clutch which runs loose on the end of a cam shaft. The other half 
of the coupling slides along the cam shaft and is thrown into mesh by a hand operated 
lever on the left-hand side of the machine. All the cams are mounted on the one shaft. 

The welding electrodes are clamped in position on the front of the secondary casting. 
The transformer is pivoted, and can be tilted forward by a cam to bring the electrodes 
into contact with the link. A spring pulls it back after the welding of the link has 
taken place. Each of the upsetting tools is operated by a separate two-step cam.^ 
The first step closes the ends of the link to make contact and the second upsets the 
weld. When welding is complete the tools are opened again by springs. There are 
swaging tools which are closed by cams and opened by a spring. Both sets of tools may 
be adjusted. There is also a trimming tool for cutting off the fins which are left on 
the links after swaging. It is operated through a bell crank lever actuated by a cam. 

At the extreme right of the machine is the feed cam, which acts upon a lever. At 
the feeding end of this lever is a pawl which engages the links of the chain on the feed 
stroke, and slides over them on the return stroke. The length of the feed is adjusted 
in the cam. When being welded the chain rests upon an adjustable saddle guide. 

At the back of the machine is fitted a cam-operated automatic switch. A main 
switch and a fuse box are fitted on the left-hand side of the machine. For regulating 
the speed of heating at the weld, there is a 5-speed plug box at the back of the trans- 
former. The front part of the secondary is fitted with four nipples to provide for the 
cooling water circulation. 

A fuU cycle of operations is as follows : (1) The automatic switch opens ; (2) the 
weld is upset ; (3) the transformer tilts back ; (4) the weld is swaged, and the upsetting 
tools release the link ; (5) the feed draws the next link into position ; (6) the fins are 
trimmed off the link which enters the trimming-box ; (7) the upsetting tools close 
the joint in the new link ; (8) the transformer tilts forward and causes the electrodes 
to make contact at each side of the joint ; (9) the automatic switch closes and heating 
commences ; and (10) the clutch is automatically thrown out. 


The company affirms that seam welding, the progress of which has been consider- 
ably delayed owing to the war, but which is now being quickly developed, is un- 
doubtedly the cheapest and best method of jointing thin material up to, say, No. 18 
S.W.G. in thickness. With it welds can be made at the rate of 5 ft. per minute on 
thin material — say. No. 33 S.W.G. — ^and at the rate of 1 J ft. per minute with No. 18 
S.W.G., with a current consumption of one and fieven units per 100 ft. respectively. 
A limitation of seam welding, however, is that, though there are cases in which a 
machine may be used for welding different kinds of work, generally speaking, each 
particular kind of work requires its own machine. Hence there are many different 
types which, though they work on the same general principle, and have many parts 
and motions in common, yet have their own distinctive features. It should be said, 
though, that in many cases the variation is only in the form of the electrodes and of 
the arms carrying them, the bodies of the machines remaining the same. A feature, 
however, which we beUeve is common to all the seam welders made by the company, 
is that the operator can sit down while working, it being recommended that the stool 
should be about 22 in. high, and placed in such a position that each of two pedals 
may be easily reached. The left pedal operates the top electrode arm, and it applies 
pressure for congoUdating the weld. It also automatically switches on the current 
when the pressure applied reaches the required amount. The right-hand pedal, on 
the other hand, controls the driving clutch and thus causes the electrodes to revolve. 
There are other points which all the machines have in common. For instance, the 
electrodes may be run at different speeds by means of cone pulleys, and it is also 
possible to obtain four different heating speeds by means of a plug box. We illustrate 
three examples of these machines in Figs. 29, 30, and 31. 

Figs. 29 and 31 show two views of what is known as Welder No. 60, one of them — 
that given in Fig. 29 — ^provided with a circular seaming attachment, and the other 
with 20 in. longitudinal stakes. The body of the machine shown in the two views is 
the flame. It is only the electrodes that are altered. Machines of this type are intended 
for a large number of purposes, including the production of paint drums, canisters, 
kettles, oil-cans, milk-cans, etc. In each case the electrodes consist of copper rollers 
which are driven at definite speeds by means of chain gearing, provision being made 
to vary the speeds to suit the material being welded. Special bearings are provided 
so that the passage of the current through them does not interfere with their mechanical 
function, and they are kept cool by a water circulation system. The pressure applied 
to the rollers is adjusted by means of a compression spring at the back end of the 
upper electrode. The machhies are power driven and require about a half horse-power. 
The primary winding has four tappings so that four welding currents are obtainable. 

Fig. 30 shows Welder No. 61, a machine which is very similar to the foregoing, 
and which is especially designed for welding the spouts on to kettles. Its general design 
and method of working are so evident from the illustration that no description appears 
to be necessary. It is intended to weld iron or mild steel sheet up to ^ in. added 

In all cases the article to be welded is placed on the bottom electrode in such a 
position that when the top electrode is brought down, by depressing the left pedal, 
it makes contact exactly where it is desired to make the weld. In some cases the 
article is clamped when in its correct position. As soon as contact between the upper 
electrode and the work is made, further pressure is exerted to switch on the current, 
and at the same time the right pedal is depressed so as to set the electrodes revolving 
for the purpose of carrying the work around the bottom electrode. In some cases the 
work is actually taken round with the lower electrode. 


The company makes spot welders in five sizes, according to the thickness to be 
welded. Standard machines are made for welding the following added thicknesses : 
0.16 in., I in., J in., | in., and | in., the ma^nmmn current required varying from 6 to 
30 kilowatts. For different classes of work a large number of different designs is 
made. For instance, the arms, or stakes, as the company calls them, are made in 
lengths varying from 9 in. to 36 in. Fig. 32 shows a group of what are known 
as No. 13 spot welders fitted with different length stakes. These particular machines 
are intended tor the welding of iron, or mild steel sheets up to J in, i^^regate thick- 

Fia. 32.— Group of B.I.W. No. 13 Spot Weldeia. 

nesB — though they may, we gather, be used intermittently for sheets up to gin. 
aggregate thickness. The top stake is fitted with a swivel joint, and the bottom stake 
can be fixed in any desired position on the face plate. The welder is foot operated. 
Depressing the pedal brings down the top electrode into contact and switches on the 
current. Directly welding temperature is reached, additional pressure on the pedal 
cuts off the current while still maintaining pressure on the weld and consolidating 
it. The electrode tips are water-cooled and are renewable. By means of four tapping 
plugs, four heating speeds for varyii^ thickness of work are obtainable. 

Some of the spot welders which the firm makes are operated by hand-lever, and 


some — the larger sizes — by means of a spoked wheel. In such machines two operators 
are required. In some machines, too, the current is cut on and off by hand, the switch 
not being interlocked with the pressure-applying system. 

In all types and sizes the method of working is practically identical, and as we have 
already described it, with, of course, slight variations necessitated by the differences 
in arrangement to which we have already alluded. The actual time occupied in 
welding may vary from a fraction of a second in the case of the thinnest material 
up to, say, 40 seconds ^with the thickest metal. The current consumption may 
similarly vary from, say, one-quarter of a unit for 1000 welds with thin sheets up to 
330 units per 1000 for the heaviest material. The speed of working varies, of course, 
with the type of machine, the thickness of material, and the personal equation of the 
operator. In one case, we understand, an operator using a small machine has made 
as many as 12,000 welds during a T^-hour day. On larger machines the output is, of 
course, less. 

As giving some idea of the different kinds of articles which are being successfully 
welded by means of this company's spot welders M'e append the following list : 
Fittings to enamelled hollow-ware ; tanks ; pulleys ; straight sheet iron tubes, and 
elbows ; spades, shovels and trowels ; automobile and bicjxle parts,' mud -guards, 
bonnets, fittings to chassis, etc. ; machinery guards ; sheets and expanded metal to 
flat and profile iron for shelves, lockers, etc. ; agricultural machine parts ; fuse parts ; 
fan propellers ; tin ware ; lattice for reinforced concrete, etc. 



In an earlier chapter we explained that, for arc welding, the only absolutely essential 
accessory, in addition to the electrode and holder, was a resistance to go in series 
with the arc. The only other thing necessary is a supply of current at a suitable 
voltage. Some consider it desirable to have an ammeter in the circuit, but in many 
cases instruments of all kinds are dispensed with for the actual welding circuits. Now, 
although the absolute essentials are only as set out above, yet it is highly probable 
that in practically every case some additional arrangements beyond those actually 
existing in any given works will have to be made in order to obtain current of the 
desired character and pressure. In the large majority of cases in this country, at any 
rate, direct current is used for arc welding. If the available electric supply be alter- 
nating, there will have to be a motor generator to convert the energy into direct 
current. If, again, the available supply be direct current, but of a voltage in excess 
of that required for the welding circuit, it will also be necessary to have a motor 
generator or a rotary motor converter to reduce the pressure. Many firm^ make motor 
generators which are suitable for arc welding circuits, and among them we may 
mention the following : The British Electric Plant Co., Limited, 78 St. Vincent 
Street, Glasgow ; The British Westinghouse Electric and Manufacturing Co., Limited, 
Traflford Park, Manchester ; Crompton and Co., Limited, Chelmsford ; The Electric 
Welding Co., Limited, 28 Basinghall Street, London, E.C. 2 ; J. H. Holmes and Co., 
Newcastle-on-Tyne ; The Lancashire Dynamo and Motor Co., Limited, Traflford Park, 
Manchester ; Mather and Piatt,* Limited, Manchester ; Mavor and Coulson, Limited, of 
Glasgow ; Bruce Peebles and Co., Limited, East Pilton, Edinburgh ; The Phoenix 
Dynamo Manufacturing Co., Limited, Bradford ; The Premier Electric Welding Co., 
Royal Liver Building, Liverpool ; The Rees Roturbo Manufacturing Co., Limited, 
Wolverhampton. Some of these firms make motor generators for converting alter- 
nating into direct current ; some machines for reducing direct current voltage, while 
some of them make both types of machines. All the machines can be used in aro 
welding, though in some cases they are simply combinations of standard motors and 
dynamos made by the various firms. In other cases, however, they embrace machines 
specially designed to produce certain specific effects, and to some of the latter we 
shall refer in what follows. Some of the firms, too, are prepared to supply portable 
self-contained arc welding generating sets which can be readily moved about from 
place to place as may be required, and are able to start producing current at a moment's 
notice. We shall have occasion to refer to machines of that type also. 

First of all we may deal with some of the apparatus supplied by the British 
Westinghouse Company. A business-like self-contained arc welding equipment, 
embodying some of this firm's machines, is shown in Fig. 33. The plant, which, as will 
be observed, is mounted on the platform of a motor lorry, and which is ordinarily 
protected by a covering which can be lifted oflf when degired, as shown in the 
engraving, was built to the order of The British Arc Welding Company (Bristol 
Channel), Limited, of Cardiff, a firm which concerns itself largely with shipbuilding 

E.W. B 



and other repair work. The generator, which is of 26 kilowatts normal output, and 
which is designed to permit of working multiple arcs in parallel and intermittently 
when using either carbon or metallic electrodes, is driven by a petrol engine designed 
to develop some 55 horse-power when running at a speed of 1160 revolutions per 
minute. The generator is intended to permit of Satisfactory operation over a con- 
siderable regulation curve, thus allowing a variety of work to be carried out economi- 
cally. The direct coupled exciter also provides current for lighting, and for driving 
portable grinding or driUing machines. 

This firm builds machines with a special winding which has for its object the doing 
away with the necessity for using a resistance in the welding circuit or, in other words, 
so controlling the working of the machine that injurious rushes of current are impos- 
sible. The winding which, we understand, was designed by the company in co- 
operation with Mr. H. S. Marquand, is shown diagrammatically in Fig. 34. There 

To Article 
to be < 



p = 


Commutating Poles 
Series iVinding 
Shunt Winding 
Separately Excited Winding 

Voltage Regulator 
connected In Shunt 

'The Enoincer" 

.Welding Current Regulator 
Connected in separate 
excitation circuit 

SwAin So. 

Fig. 34. — Diagram of Connections of Westinghouse Generator for Arc Welding. 

ajre three separate windings, or four, if that of the commutating poles be taken into 
account. The series and commutating pole windings — ^A and P respectively — ^are in 
series with one another, and, of course, with the armature. The shunt winding B is 
connected not across the armature alone, but across the terminals of the msu^hine, and 
hence outside the series and commutating pole windings. It has an adjustable resist- 
ance in series with it. There is a further winding C which is separately excited from a 
constant voltage exciter, and which also has an adjustable resistance in series with it. 
The windings B and C assist one another while winding A opposes both of them. When 
considering the effect of these three windings the demagnetising effect of armature 
current must not be lost sight of. Let us suppose that the machine is running at fuU 
speed, but with open circuit. Then the series winding is practically inoperative, since 
it is only carrying the comparatively negUgible current required to energise the shunt 
winding B. On the other hand, the latter winding is fully effective, and the winding 
C is energised to give in conjunction with winding B the desired initial voltage on the 
armature. -If now the electrode, that is to say, one terminal of the generator, be 



brought into contact with the '' work," which is connected to the other terminal, the 
tendency is, naturally, for a heavy current to flow through the armature and the 
series winding A. As soon as current begins to flow, however, the demagnetising 
effects of the series winding and of the armature reaction tend to reduce the armature 
voltage, the amount of reduction depending, of course, on the relative strength of the 
series field and the combined shunt and separately excited windings. But this is 
not all that happens, for, since the generator terminals are short circuited, the shunt 
winding B is also short circuited, and becomes non-effective. Hence, the separately 
excited winding C is then opposed by the series winding A and the armature reaction, 
and matters can be so arranged that the short circuit current will be actually less than 
the normal full load current of the machine. In practice, of course, the actual time 
during which the machine id short circuited is only under ordinary working conditions 
for a fraction of a second, but even if by accident the short circuit period is prolonged, 
no harm can happen to the machine, and in working ne steadying resistance is 
required in series with the arc. As the makers point out, this means a considerable 
reduction in the power consumed, since all the C^B losses in the resistance are 
avoided. It is of interest to see to what these losses amount. 

Let us suppose, say, that a welder is taking current from a 100- volt generator. 
If he be using a metal electrode, the current may be between 50 and 200 amperes. Take 
it that the average is 125 amperes. The voltage across the arc may be, say, from 25 
to 40. Let ud take the best case as far as losses are concerned, and imagine it to be 40. 
That means that the resistance is absorbing 60 volts. The energy being consumed in 
the resistance is, therefore, 125 by 60, or 7500 watts, while the energy in the arc which 
is doing the welding is only 1 25 by 40, or 5000 watts. More than half of the total energy 
in the circuit is therefore absorbed in the resistance. Of course, this does not quite tell 
the whole tale, for the energy consumed in the exciter of the protected machine has 
to be taken into account, but the loss in that machine id small compared with the loss 
in the resistance. Moreover, the loss is, of course, greater if a higher voltage than 100 
be used. With a carbon arc, if we take the figures of the Steel Barrel Company, 
limited, of Uxbridge, to which reference has been made in an earlier chapter, the 
initial voltage is 90, and the voltage across the arc from 50 to 55. Taking the higher 
figure, as was done in the case of the metal arc, the voltage absorbed is 35, so that 
whereas the arc, with a current of 300 amperes, absorbs 300 by 55, or 16,600 watts, 
the resistance loss is only represented by 35 by 300, or 10,500 watts. 

Thus in the two cases we have taken, whereas with metal electrodes the ratio of 
energy wasted to energy usefully employed in the arc may be as 7.5 to 5, with a carbon 
arc the ratio is as 10.5 to 16.5. In both instances the losses are heavy, but proportion- 
ately the carbon arc system appears to be the less wasteful. 

Let us attempt to get some idea of what this waste means in pounds, shiUings and 
pence. Suppose for the sake of argument that a carbon arc welder works eight hours a 
day, and that he is actually using his arc for only half that time, say, four hours for 
five days a week and two hours on Saturdays, a total of twenty -two hours per week. 
During that time the energy waited in the resistance will be 10,500x22=231,000 
watt-hours, or 231 Board of Trade units. Taking the cost of the energy at Id. per 
unit, there is, hence, a loss of nearly £1 per week, or, say, £50 per year for each welder. 
These figures become worse the greater proportion of his working time the operator is 
using his arc, and if he were actually welding during the whole of his time the annual 
loss per man woald, on the basis of calculation we have taken, be over £100. 

By means of the two adjustable resistances in the Westinghouse-Marquand 
arrangements, the current flowing in the arc can be varied within certain limits, so 













that the same machine can be used for different kiiidB of welding work calling for 
CQirents of different strengths. In addition to this means of control, ve understand 
(hat slterstion of the brush positions affordb a further means of regulation. The com- 
mutation is said to be sparkless tmder practically any brush position. Sets of this 
type have been successfiiUy arranged on boats, so that work can be carried on from 
the water, an arrangement which in ship repairing is sometimes most convenient. The 
generators are rainproof, since they may frequently be called upon to run continuously 
for long periods under conditions of weather and position which may be anything 
but good. 

Going back to the plant illustrated in Fig. 33, it may be mentioned that each lorry 
is provided with 550 yards of cable, so that current can be conveyed for that distance 
away from the point at which the vehicle 
is stationed. The cable has cab-tyre 
sheathing and is carried on rollers in 
damp ■ proof boxes mounted on the 
chassis. The lorry can carry some 150 
gallons of petrol in its tanks. The 
tank, which is seen mounted above the 
generator, is kept full by pumping up 
from a larger tank arranged beneath 
the chassis. 

Another self-contained welding set 

of a similar character is Bupplied by the 

Premier Electric Welding Company, of 

Royal Liver Building, Liverpool. It is 

shown in Figs. 35, 36 and 37. It also 

can be mounted on a lorry or in a special 

hut of its own, which can be slung on 

shipboard and lowered on to the deck 

V)r into a hold (see Fig. 36). The set was 

specially designed for use with the metal 

arc process. It consists of a 10-brake- 

horse-power petrol engine, direct coupled 

to a 4.3-kilowatt generator with a sepa- 

FiG. 36.-Premier Welding Set in Cabin. rate exciting dynamo connected to the 

some shaft and mounted on a common 

bed-plate. The main generator has three windings— a series winding, a shunt winding 

connected across the armature, and a separately excited winding. The two latter 

assist one another, while the series winding opposes both of them. The combined 

effect of the shunt and separately excited winding produce a field strength which 

determines the voltage of the generator before the welding arc is established. When 

the arc is struck, however, current flows through the armature and the series 

winding, with the result that the strength of the field is reduced and the voltage 

in consequence lowered. When the combined effeot of the series winding and the 

armature reaction is sufficient to balance the counter-effect of the separately excited 

winding, the resistance drop of the external winding of the generator will then 

determine the voltage of the generator terminals. The makers point out that 

with a separate exciter maintaining a constant current, the armature current of 

the main generator must necessarily remain constant in order to keep the armature 

reactance and the differential series field constant. This gives a machine with an 


external current characteriHtic which is nearly constant within the limitation of 
the welding voltage. 

In opere>tion a reactance is used in series with the arc on the negative leg of the 
generator. This reactance has four terminals, which are so connected to the two ter- 
minals of the welding circuit and to the terminals of a three-pole double-throw &^«-itch 
that in one position of the latter the two coils of the reactance are connected in parallel, 
whilst in the other position they are in series with the generator and the welding arc. 
It is claimed that a reactance and switch so arranged in connection with a generator 
wound in the manner above described enable an operator, without having to change 
the generator ipeed or any of the connections, to arrange simply by throwing the 
switch over in one direction or the other to provide either a small current for welding 
small articles and building up thin and worn plates, or a larger current for use in con- 
nection with heavier work. 

Fig. 38.— LaucaBliire Dyiiamo and Molor Compauy's Arc Welding Set. 

Another differentially wound generator, specially intended for electric arc welding 
work, is made by the Lancac^hire Dynamo and Motor Company, Limited, of Trafford 
Park, Manchester. It is shown coupled direct on the same bed-plate to its motor in 
Fig. 38, while a diagram of the windings of the two machines is given in Fig. 39. As 
will be observed, the arrangement of the winding is different in several particulars 
from the windings oi the machines which have been described above. There is a shunt 
winding A which is connected across the armature of the generator, and has a regulator 
in circuit with it, and a series winding which is in the are welding circuit, and there 
is a third separately excited winding B which, however, is arranged in series with the 
field winding of the motor across the supply mains. 

In describing the arrangement the makers write as follows ; " The arc may be 
struck across the armature of the generator without any resistance in circuit, and 
the plant is capable of running for short periodp with the armature of the generator 
short circuited." The windings A and B tend to assist one another, while the winding 
C opposes both, and hence tends to lower the generator voltage. No series resistance 



is used in the welding circuit. At the moment of touching the work with the electrode 
there is a tendency for an excessive current to flow in the generator circuit. This 
current, of course, flows through the series winding, and is limited to the value which» 
will demagnetise the generator field by an amount sufficient to prevent any larger 
current from flowing. Similarly, when an arc is struck, the resistance of the circuit is 
increased, and the current tends to fall. This tendency is compensated for by a 
diminished demagnetising effect of the C winding, whereupon the generator voltage 
is increased to an amount which will permit of the flow of the desired current." 











r< Shunt 
< Regulator 

\ Supply 
I to A re 

• •• 

C Winding 


'the EnQineer" Swain SO 

Fig. 39. — Connections of Lancashire Dynamo and Motor Company's Arc Welding Set. 



As an example of apparatus particularly suitable for arc welding we may at this point 
refer once more to the diflEerential electro-magnetic clutch devised and patented by 
Messrs. Walter L. Davies and Alfred Soames, both of Faraday House, 66 Southamp- 
ton Row, Holborn, London, which was described at some length in the issue of 
The Engineer of January 25th, 1918. The advantages claimed for this device, for 
which the Equipment and Engineering Company, has acquired the exclusive licence 
in the United Kingdom, are, as far as electric arc welding is concerned, as follows : — 

(1) Any dynamo giving Sufficient power and driven from any suitable source of 
power may be used, its characteristics being so altered as to make it suitable for the 

(2) The maximum current to be taken can be set at any value within the limits 
of the dynamo. 

(3) When the apparatus has been set for any desired current the latter will remain 
practically constant from short circuit to the maximum length of arc that the volts 
will maintain. There is no necessity to " snatch at " the arc when striking it. The 
electrode can be rubbed on the work, and the arc drawn cut as desired. A short arc 
does not produce a rise of current and consequent burning. The operator is thus 
enabled to give his full attention to the work, since the arc cannot '" run away from 
him." There are no jumps or breaks in the current. 

(4) No resistance is required in the welding circuit, and consequently the whole of 
the watts generated are used upon the work. 

The clutch is shown in Fig. 40. It may be placed either between 'a motor and a 
generator, as shown in Fig. 41, or between a steam engine or other prime mover and 
a generator. A is the driving shaft and B the driven shaft, which is coupled direct 
to a separately excited dynamo. Keyed to the shaft B is a coupler D, which is furnished 
with driving-pins H. Housed in the coupler is a Skefko bearing C, which receives 
the end of the shaft A. Near the end of the shaft A is another Skefko bearing E, which 
is housed in a disc G, to the periphery of which is bolted a cast-iron ring F, the disc 
G and the ring F forming the transmission momber of the clutch. In the disc G, 
-arranged at similar intervals to the pins H on the coupler D, are holes in which are 
inserted leather rings J, the latter also being pierced with central holes to accommodate 
the pins H. The disc G and the ring F are free to float and move to the extent per- 
mitted by the Skefko bearing. The body of the clutch is indicated by the letter K, 
and the magnetic circuit is completed by the ring L. 

There are three windings. A, B and C. Fig. 42 is a diagram showing the basic 
principle of the application of the clutch to a motor generator when the apparatus 
is used for electric arc welding, and Fig. 43 shows the electrical connections. In both 
these cases the clutch is, of course, only shown diagrammatically. Coils A and C are 
made up of many turns of fine wire, while coil B consists of comparatively few turns 


of thick wire. No part of the clutch moves longtitudinally, bo that its reluctance 
remainB constant and the variation of the magnetism produced by the coils varies the 
pressure of the clutch members against one another only. 

The action of the clutch is as follows : The coil A is energised either from the 
mains which supply current to the motor, or, if an alternating current motor be used, 
from a small exciter on the driving shaft. The coil is so proportioned as to produce 
an initial pressure of the clutch members, which is more than sufficient to transmit 
the load. Coil B is connected in series with the armature of the dynamo on the driven 
shaft, and its polarity is such as to counteract the eflect produced by the coil A. That 
is to say, as the current in the coil B increases, as it does when the load on the 
driven machine increases, the pressure of the clutch members against one another 
decreases, until a point is reached when the pressure is only just sufficient to transmit 

Fig. 40.— The Davies-Soames— Day sohms— Electro- Mtgnetic Differential Clutch. 

the load without slippii^. If that point be passed the clutch slips. It is apparent, 
therefore, that the current taken from the dynamo cannot pass a given maximum 
even if the machine be short circuited. The particular maximum to be reached can 
be arranged by adjusting the excitation of the coil A with a rheostat. 

The coil C, which, like coil B, opposes and tends to weaken the effect produced 
by the coil A, is connected across the brushes of the driven dynamo. In a separately 
excited dynamo the torque required to drive it at any speed is proportional 
to the current taken from the dynamo, irrespective of the speed and volts. If a con- 
stant current be taken, it requires a constant torque to drive it at any speed in a fixed 
field. It is to counteract the variation in the coefficient of friction that the coil C was 
added to the apparatus. On open circuit the coil A gives an initial pressure or attrac- 
tion between the members of the clutch. The coil C, which, as explained above, is 
Arranged across the brushes of the dynamo, reduces the pressure produced by A. 
If current be now taken from the dynamo the coil B still further reduces the pressure 



or attraction between the clutch members until a point is reached when the clutch 
slips. When slip occurs, however, the volts of the driven dynamo decrease, and, 
consequently, the effect of the coil C is weakened, which is equivalent to increasing 
the effect of A. By this means the slip is prevented from increasing unduly. Should 
the current taken from the dynamo tend to increase, the action of the series coil B 
would decrease the pressure attraction between the two clutch members. As a net 
result, therefore, the coils B and C adjust the pressure, so that no matter what is done 
with the arc, the current remains practically constant. By suitably proportioning 
the ampere turns in coils A and C, which may be done by means of rheostats in series 
with them, any desired characteristics, within reasonable limits, may be obtained. 

Tmc Encinccr" 

Fig. 42. — Davies-Soanies Chitch arranged for Arc Welding. 

The clutch, therefore, can be used for driving a dynamo the current from which is 
being used for light or heavy work. 

We may add that since our description, of which the foregoing is an abbreviation, 
was published, and, indeed, since the earlier chapters of this book were penned, Messrs. 
Davies and Soames have made sundry modifications of their device, which, they 
claim, have increased its efficiency. For example, they have altered the shape of the 
magnetic circuit of the clutch, with the result that there is less leakage of lines of force 
and a higher efficiency is arrived at. Moreover, they have succeeded in reducing the 
electro-motive force of the welding circuit to as low a figure as 33 volts. We have 
watched the apparatus in operation with that voltage, and observed that, as might be 
expected with such a low pressure, it was quite impossible to have a long arc, with 
its accompanying disadvantages. At the least attempt to draw out the arc beyond 
a certain point the circuit was interrupted. On the other hand, a skilled operator 
appeared to have no difficulty whatever in maintaining an arc and in doing excellent 



welding with a current that only yaried within Bmall Hmitg, the wcvk prodooed 
of an even charaeter. A coated metalfic electrode was, of coarse, being used. It 
should be mentioned, in conclo8ion,that Messrs. Davies and Soames now use a reactance 
m the welding drcuit. 

While on the subject of special windings, attention may be drawn to an article 
which appeared in the December, 1918, issue of the Am^can General Electric Review^ 
and was entitled "The Constant-Energy Arc-Welding Set.'' The article was written 
by Mr. P. O. Noble, of the Direct Current Engineering Department of the General 
Electric Company of Schenectady. The author takes for his text that '* all will agree 

Dynamo JSL 

Spindle \ 

Separately Excited 
From Moto r Circtnt or 

small Exciter tn case 
of Alternating Current 

When the polarity of Coil 
A is -i-.the polarity of B 
and C are — 
Bt C »ork against A. 

Dy^ Field 

Separately Excited From same source as Coil A. 

A is constant, a Fall oF Volts across C. Increases pressure oF Clutch 

a rise oF Current in B. Decreases . •* 

Swain Sc. 

Fia. 43. — Diagram of Connections of Davies-Soames Clutch. 

that the ideal condition for a homogeneous weld on a given piece of work is obtained 
when the voltage and the current at the arc are constant," a condition which is, he 
maintains, impossible with the manually controlled arc. Hence the raison cTetre of 
the machine which he describes, the principal advantage claimed for which is that it 
facilitates the maintenance of a short arc and makes it difficult to obtain a long one. 
Mr. Noble explains that the rate of consumption of electrode material is proportional 
to the current in the arc and is independent of the voltage across the arc. ^^ In the 
constant-energy system," he continues, '' when the operator shortens the arc, the 
current increases and the rate of wire consumption increases, thereby tending to bring 
the arc length back to normal. If he lengthens the arc, the current decreased and the 



rate of wire consumption decreases. This automatic action tends to maintain the arc 
at the proper length. If the arc is unduly lengthened, the current will decrease until 
the electrode does not fuse properly and the arc will go out. In the constant-current 
system, the rate of wire consumption is constant, and there is no such corrective effect 
as that just described. This system also makes possible the maintenance of a long arc 
with the consequent deposition of a large amount of highly oxidised and porous metal. 



Comm Motor 
Field Armature 



The Enoincer" 

Fig. 44. — Electric Connections of Constant-Energy Balancer Set 

Swain Sfr 

The oxidisation is evidenced by the fact that with a long arc there is present, in addition 
to the black oxide of iron — ^Pg O4 — which is present on all bare wire welds, an abun- 
dance of red oxide of iron — Pcg O3 — showing that the metal is more highly oxidised 
in passing through the arc." 

The constant-energy balancer set consists of a direct-current motor mechanically 
connected to a direct-current generator. The electrical connections of the two machines 
are shown in Fig. 44. It will be observed that, in addition to being mechanically con- 


nected, the motor and generator are ako electrically connected in series across a supply 
circuit of constant potential. One terminal of the welding circuit is taken from the 
connection between the two armatures, and the other terminal from the positive 
line. The polarities of the motor and generator are so arranged that the armature 
currents of the two machines add together in the welding circuit, which, Mr. Noble 
remarks, makes possible an approximate 50 per cent, reduction in the size of the units 
involved. No series resistance is used in the motor, generator, or welding circuits, 
the difference between the supply voltage and the arc voltage being absorbed by the 
motor. This, adds Mr. Noble, '' results in saving the power which is usually wasted 
in rheostats. The ratio of the energy at the arc to the input to the set is 66 per cent. 
With the ordinary set of corresponding capacity using series regulating resistance, this 
ratio is approximately 22 per cent." 

The initial voltage for striking the arc, and the average current for the particular 
work in hand, may be regulated at will by means of the motor and generator field 
rheostats, and by means of the switch which changes the number of effective turns in 
the differential series field winding of the generator. Changes of current may be 
obtained, it is stated, from 40 to 150 amperes by very small increments. After the 
arc is established, the variations in current and voltage at the arc with changes in 
the relative positions of electrode and work, follow, approximately, a constant energy 
law. The regulation is entirely inherent in the set itself, and is accomplished without 
the use of energy-consuming resistances or vibrating regulators. 

Mention may also be made of a rotary motor converter or self-contained rotary 
transformer, which is made by the Equipment and Engineering Company, of 2 and 3 
Norfolk Street, Strand, London, W.C. 2. The machine has a single field and a single 
armature, the latter being furnished with a double commutator. The motor side is 
arranged to absorb current at 500 volts, and the generator side to deliver current at 
from 80 to 100 volts. The current available for the welding arc is 250 amperes for 
steady working, but it can be raised by 50 per cent, or to 375 amperes for short periods. 
The machine is of the semi-enclosed type, and is fitted with carbon brushes. It is 
especiaUy intended for use in traction work in which the working voltage is in the 
neighbourhood of 500. 



Large numbers of portable generating sets designed for arc welding in works and 
shipyards have been supplied by J. H. Holmes and Co., of Neweastle-upon-Tyne. The 
particular arrangement adopted by this firm is shown in Fig. 45.* The generating 
plant consists of a four-cylinder petrol engine of 28 brake-horse-power running at 
850 revolutions per minute, coupled to a direct current compound wound dynamo 
having an output of 250 amperes at 66 volts. This voltage, it will be noticed, is con- 
siderably lower than that employed in some cases of arc welding. The plant is 
mounted on a combined bed-plate, and has a special framework which enables the 
equipment to be easily lifted and moved from place to place. On the framework is 
also fixed a radiator for cooling the circulating water, the air being drawn through it 
by means of a fan driven from the dynamo shaft. There is also mounted on the frame- 
work a small switchboard carrying a single pole switch, double pole fuses, shunt 
regulator, volt meter and ammeter. The special resistance used in conjunction with 
this plant is shown in Fig. 46. It has four switches mounted on a sjiate base in front 
of the resistance. The four switches have capacities of 100, 75, 60 and 25 amperes 
respectively. The resistances are so connected up that it is possible by different 
combinations of the switches to obtain currents of from 25 up to 260 amperes in ten 
steps of 25 amperes each. The resistance is built up of cast iron grids, having the 
firm's patented stiffening arrangement, which strengthens the grid and, it is claimed, 
makes it absolutely rigid, thus preventing breakages due to movement or vibration. 
Numerous self-propelling arc welding equipments have been supplied to ship- 
building and arc welding companies in London, Belfast, Cardiff, Glasgow, Liverpool, 
South Shields, etc., by Tilling-Stevens, Limited, of Victoria Works, Maidstone. This 
company has developed three types of equipment, in each of which the vehicle is self- 
propelling, and is driven by the well-known petrol-electric combination which is 
associated with the name of the firm. In the first case the vehicle has an ordinary 
lorry type body on which, at the back of the driver's cab, is a compartment for 
containing the control gear and equipment, thus leaving a large proportion of the 
platform free for ordinary carrying purposes. In another form, which we gather is 
the more popular of the two, the vehicle is provided with a box-type body, which 
serves as a workshop and a mess-room for the operators, as well as for containing the 
welding gear and accessories. This type of vehicle is frequently fitted with a petrol- 
driven air compressor for driving pneumatic tools. Both of these equipments are 
intended for supplying the current for one arc only. In the third equipment arrange- 
ments are made for supplying two arcs, one of which derives its current, as in the case 
of the first two, from the main dynamo, which also serves to propel the vehicle through 

* In a letter to the Editor of The Engineer dated April 23rd, 1919, and published in the issue of 
that Journal of May 2nd, Mr. J. E. Weyman of Newcastle-upon-Tyne pointed out that the illustration 
shown in Fig. 45 appeared to represent a set which he designed and supplied to the British Arc Welding Co., 
Ltd., and of which, with some modification, he had since supplied some eight further examples, for all of 
which Messrs. J. H. Holmes and Co. supplied him with the electrical outfit. 

E.W. F 


the intennediary of a motor, while the other derives ita current from a small separate 
direct-coupled petrol-electric generating set. Exterior and interior views of one of 
the second type of equipments are given in Eigs. 47, 48 and 49, while an exterior view 
of another design is given in Fig. 60. The chassis in all three coses is the same as that 
of the firm's 40 horse-power standard vehicles. The prime mover is a 40 horse-power 
petrol engine, which is direct coupled to an electric generator, designed to give 80 to 
100 volts, and compound wound to give a level or slightly falling voltage curve. For 
propelling the vehicle tMs machine, of course, supplies current to the motor, which in 
turn drives the back axle through a cardan shaft and worm gearing. The control of the 

Fro. 46.— Holmes Beeietance Frame for Arc Welding. 

vehicle when travelling on the road is effected by means of the engine throttle and a 
resistance, which varies the strength of the generator and motor fields, the entire rai^ 
of speed being controlled by the throttle pedal and resistance lever. The body is of the 
box type, and is provided on the near side with hinged doors for lighting and ventila- 
tion, and on the off side with one window and louvres. The driver's cab is boarded 
up on the off side, and there is a lunged door on the near side. A sheet-steel weather 
screen is also provided. At the back of the cab folding doors give the driver access 
to the welding control gear, so that he can manipulate it from bis seat. In the interior 
of the body, on the near side, there is a-locker seat extending for about half the length, 
and a bench fitted with a vice, and provision underneath to contain cable for con- 
veying current to the welder — and pneumatic hose if a compressor is fitted — occapies 
the remaining space on that side of the body. 


The switchgear and resistances for the welding circuit are fitted in the front 
portion of the body on the off side. The oablee from this gear run down through the 
floor to terminals on either side of the body. From these terminals the cables can 
be taken to where the are is required. The welding leads are supplied in convenient 
lengths, which can be thimbled together as required, and junction boxes are provided 
for the insulation and protection of temporary connections. 

The plant is designed to weld mild steel plates from J in. to 2 in. thick and upwards, 
and for all descriptions of steel and iron forgings and castings. The output of the 
dynamo is from 18 to 20 kilowatts. When the car has been taken to the place where 
the repairs are to be carried out, the controller-lever on the steering column is put into 
the neutral position, and the main switch on the welding switchboard is closed. The 

Fio. 48.— Tilling-StevenB Arc Welding Lorry— Interior. 

motor is then entirely out of the circuit, and the engine and dynamo are set ready for 
generating current for welding. 

Another firm which has specialised in the manufacture of machines for supplying- 
current for arc welding is that of Crompton and Co., Limited, of Chelmsford, and it 
has turned out some large machines for this particular class of work. It makes 
machines of various types, from those designed to keep one single arc going to those 
intended to supply current to several separate welders working in parallel, each with 
his own steadying resistance. Such generators as the latter are simply compoimd 
wound, so as to give a level characteristic and to ensure stability. They do not differ 
essentially from ordinary compound-wound dynamos, and do not therefore call for 
special comment or for more detailed reference. Single welder machines, however, may 
sometimes be compound wound so as to give a faUing characteristic, that is to say, 
be oyer compounded so that increase in current tends to reduce the voltage, or they 
may have three windings, one of them energised from a separate exciter. Then, 
again, the machines may be motor -driven, the motor being either for alternating or 
direct current. All the Crompton machines intended for arc welding purposes are, so 


we nnderBtand, demgoed to withstaod overloads of 25 per cent, for two hours, 50 per 
ceat. for short periods aod 100 per cent, momeatarily. To make a provisioD of this kind 
is distinctly wise, because the dnty of a welding generator is witbont doabt extremely 
udDOdfl, since the load is s very varying quantity. la the case of a small generator, 
supplying only one welder, the variation may well amount, momentarily, to overloads 
of twice the normal, the load rising instantaneoDEtly from zero value. 

Two machines have been chosen to illustrate the special ptodocts of the Crompton 
firm. Th^ are ahown in Figs. 61 and 52. The former shows a petrol-driven self- 
ocHitained welding set, the engine, petrol tank, oil tank, radiator, generator and switch- 
board being mounted on one bed-plate, so as to be readily portable. The generator 
is otmipound wound to give any desired voltage characteristic, and is of 5 kilowatts 
capacity at 110 volts when running at 1500 revolntions per minute. The machine 

Fio. 49.— Tilling-SteveDB Ak Welding Lotty— Driver's SeaL 

shown in Fig. 52 is a motor generator of 12 kilowatts capacity at 60 volts. The motor 
is of the three-phase type, and the direct-current generator has three windings and a 
separate exciter, so that effects which have been discussed in an earlier chapter in 
connection with other machines are obtainable. 

As examples of some of the electrode holders for metal or carbon electrodes now 
being used, we may draw attention to Fig. 53, which shows two types of holder. 
That on the left is intended specially for use with metal electrodes up to No. 6 gauge, 
and in it the electrode is gripped by spring action, and there are no binding screws to 
manipulate. The heat guard is of vulcanised fibre. The holder on the right is of a 
heavier pattern, and is intended for carbon or graphite electrodes, from | in. up to j in,, 
and for currents up to 250 amperes. The heat guard is of metal. Both these holders 
are made by the Equipment and Engineering Company, which also supplies a still 
heavier holder, designed to take electrodes up to 1 in. in diameter, and to be suitable 
for currents of 400 amp^reB. 

According to Mr. A. M. Candy, as set out in a paper which be read in November, 


1918, before the Pittsburg Section of the American Inetitute of Electrical Engineers, 
the carbon holder, as used in the United States, comprieeB an aluminium rod, one 
end of which \b fitted with a connection for attaching the feeder cable, while to the 
other end is welded a tube which contains two jaws or clamps for holding the electrode. 
The portions of the clamps which are in the tube are tapered, so that when forced into 
the tube they grip the electrode firmly. The cable end of the holder is furnished with 
an insulating, heat-resisting handle, and a disc shield to protect the operator, not only 
from electric shocks but from the heat and light rays emanating from the arc. The 
electrode is of solid graphite, usually one inch in diameter and from six to eight inches 
long. The holder, which is intended for currents up to 600 amperes, is, Mr. Candy 
explains, sufficiently Ught to be handled easily sad is strong mechanically. For 
heavier currents a larger holder is necessary. The metal electrode holder as used in 

Fig. 60,— Tilling-Stevena Arc Welding Lorry. 

the United States — and we repeat here that in that country bare metal electrodes were 
much more extensively used, at all events until comparatively recently, than were 
covered electrodes, whereas the reverse is the case on this side of the Atlantic — is 
considerably Ughter than the carbon holder, and the disc shield is generally omitted 
because the heat at the arc is much less intense. The electrode is clamped in the 
holder by a earn device, which is so designed as to be capable of accommodating 
electrodes of various diameters. 

The forms of screens employed for the protection of the eyes and faces of operators 
when engi^ed in arc welding are numerous and varied. We illustrated one of them in 
an earlier chapter, and we might deal with many more if it were not that no really 
useful purpose would be served by doing so. The fact of the matter is that it does not 
matter what shape or form they take as long as they afford efficient protection. For 
sonje classes of arc welding the necessity for protection is considerably less urgent 
than in others. The long carbon arc, for instance, has a much more potent effect on the 
surfaces of the human body than have the considerably shorter arcs with which some 


types of coated metal electrodes operate. Even bo, we have seen caeee in which only 
a Bcreen held in. the hand was employed for carbon arc welding, while much more 
elaborate precautions are BOmetimes taken by coated electrode welders. It is quite 
clear, therefore, that there is no reason why the screens or shields should be of any 
particular shape, though it would be as well for factory owners to satisfy themselves 
that what is employed will, if properly used, afford the necessary protection. 

In this connection there arises the question of the best type of glass to use. It 
must be admitted that hitherto this aspect of the matter has been treated by many — 

Fio, 51.— Crompton Selt-coDlained Arc Welding Set 

though not by all — in a far too haphazard manner. Notable examples to the contrary 
were the labours of such investigators as the late Sir William Crookes, who devoted 
much careful study to the production of glass which, while affording a good view 
of the worki would protect the eyes of workers. I'he problem, however, has not, w© 
think, yet reached a definite solution, though several different kinds of " safety " 
glasses have been evolved, and many combinations of coloured glasses are in use. 

It is comparatively easy to select a combination of colours which will cut off a 
very large percentage of the harmful rays, whether of light or heat, and which will at 
the same time enable the worker to see what he is doing perfectly well when once 


the are has been struck. With many such combinations, however, it may be quite 
impossible, with ordinary illumination, to see anything at all of the work. Such 
a disability cannot be helpful, though we must admit that, when observing the opera- 

Fio. ft2.— Crompton Motor Generator Set for Arc Welding. 

Fig. 63.— E. and E, Company Elt^trode Holders. 

tione of workers using combinations of coloured glasses through which ordinary 
daylight could only be faintly seen, we have been struck by the »mall amount of 
hindrance it has been. 


It has to be remembered that it is not only the visible light rays which have to be 
guarded against ; there are also those in the infra-red and ultra-violet ends of the 
spectrum. Glass which may cut off those at one end may allow the free passage of 
those at the other. It has, however, not been beyond the powers of science to provide 
a solution to the problem. Certain glasses having greenish tints have been found to 
give excellent results, and it is rather curious that they have not been more extensively 
used in this country than has, so far as we are aware, been the case. Glass having 
tints of this character and capable of passing a large proportion of the visible rays, 
has considerable vogue amongst welders in the United States, especially that made 
by the Corning Glass Company, of Corning, New York. Gold-plated glass has also a 
remarkable capacity for stopping heat rays, but it is inferior in transparency to other 

The following summary of the effective means for eye protection against the 
various harmful radiations that are particularly 'associated with welding operation, 
forms the conclusion of an article recently written on the subject by Mr. W. S. Andrews, 
of the Consulting Engineering Department of the General Electric Company of America, 
and wiU doubtless be read with interest. It appeared in the December, 1918, number of 
that company's Review, from which we have already quoted in preceding chapters : — 

(1) ''The intense glare and flickering of the visible rays should be softened and 
toned down by suitably coloured glasses, selected by an expert and having a depth of 
colouration which shows the clearest, definition combined with sufficient obscurcUion of 
glare, which last feature can be best determined by the individual operator. 

(2) '* When infra-red rajrs are present to a dangerous degree, a tested heat-absorbing 
or heat-reflecting glass should be employed, either in combination with a suitable 
dark-coloured glass, when glaring visible light is present, or by itself in cases where 
the visible rays are not injuriously intense. 

(3) " In guarding the eye from the dangerous ultra-violet rays, it must be care- 
fully noted that ' pebble ' lenses are made from clear quartz or natural rock crystal, 
and this material being transparent to these rays offers no protection against their 
harmful features. On the other hand, ordinary clear glass is a protection against 
these rays when they are not very intense, but dark amber or dark amber-green glasses 
are absolutely protective. Glasses showing blue or violet tints should be avoided, 
excepting in certain combinations wherein they may be used to obscure other colours." 




Among the pioneers in the use of the Thomson system of electric resistanxje welding 
in this country was the Electric Welding Company, Limited, of 2S Basinghali Street, 
London, E.C. It has consequently bad a considerable amount of experience in the 
design and construction of machines working on this principle, and the result of that 
experience has been that it has arrived at the conclusion that it is wise to make 
electric welding machines amply strong, since the users of such machines are very apt 
to demand from them very considerably more than their rated capacity. Hence one 
of the distinguishing features of the machines which the company builds is robust 
construction, and since the machines are also characterised by good mechanical and 
electrical design a ghort reference to s'>me typical examples will doubtless be of 
interest. It so happens that just recently the company has introduced several new 
types of electric butt welders, and to them we propose to refer in what follows. It 
must not, however, be thought that it only constructs machines of that type, for it 
also makes spot welders, seam welders, wire welders, chain welders, and machines for 
many other purposes, aU of which it would be manifestly impossible to discuss. 

The machine illustrated in Fig. 54 is intended for butt-welding heavy steel tubes 
up to 3| in. in diameter. It will be observed that the electrode clamps for gripping the 
pipes are of the horizontal type with top opening so that the pipes to be welded may 
be dropped in from above. This construction is advantageous when the work is of a 
heavy character, or when the tube being dealt with is in the form of a close spiral as, 
for example, in the case of the coils for a refrigerating machine. The adjustment 
between the clamps to suit various diameters of pipes is obtained by means of the two 
hand wheels at the back of the machine which adjust the position of the rear clamps. 
Each of the forward or working clamps is furnished with a hand-operated cam lever 
by which the work is firmly gripped by one movement. The clamping blocks and the 
platens on which they move are cooled by water circulation. 

Heavy work such as the welding of pipes as large as 3| in. in diameter requires a 
considerable pressure properly to consolidate the weld, and in the machine under 
consideration it is applied by means of a 12-ton hydraulic jack which is operated by 
a lever. 

The machine, working as it does on the Thomson principle, embodies a static 
transformer, and the primary winding is provided with five tappings controUed by a 
selector switch for adjusting the voltage, and hence also the current in the secondary, 
winding, to suit the various sizes of tube, etc., dealt with, the smallest tube which the 
machine is designed to weld being 1^ in. in diameter. The main control switch, which 
is oil immersed and enclosed in a galvanised steel tank, is operated by the foot lever 
and wire rope, which may be seen at the front of the machine at the left-hand side. 
Machines of this type, some of which are, we understand, being used by some of the 
largest tube makers in this country, are also suited for the welding of straight rods and 

Fig. 64.— Electric Welding Co.-Butt Welder for Tubes, Rode, etc 

Fw. 65.— Electric Welding Co.-Biill Wel,i«r with Radial Weldin 


bare, angles, corner welds in rectangular frames, and other forma of work in effecting 
which it IB convenient to insert the parts to be welded from the top. 
tfts" A machine similar to the foregoing in size, but differing from it in the fact that the 
clamps are of the horizontaLoblic|ue type, is shown in Fig. 55. In it the welding blocks 
are arranged in radial fashion, and are operated by screws and hand wheels. Tlus 
form of clamp is especially suitable for the welding of rings, tyres and other endless 
shapes of work, and the particular machine illustrated is designed to deal with cross- 
sectional areas up to six square inches. Adjustable springs are provided so that the 
applied pressure may not exceed a predetermined amount. 

Fio. 56. — Electric Welding Co.— Butt Welder for Miacellaneous Repair Work. 

A welder specially designed for miscellaneous repair work, which has been adopted 
by several British railway companies for renewing the iron work of rolUng stock, is 
shown in Fig. 66. The four clamps, or jaws, are set at an angle and are adjustable by 
hand wheels. Each has its own water circulation system. The pressure is appUed, as 
in the case of the first two machines, by means of a hydrauUc cyhnder. This type of 
machine has proved very serviceable in many kinds of railway material repair work. 
For example, the truM rods, brake rods, draw-bar rods, and similar parts of the 
underframing of coaches, are frequently found to be corroded and worn, while the 
forgings at their ends may still be in good condition. By cutting away the worn rods 
and welding on new ones the cost of new forgings is saved, and the resulting economy 
is found to be quite considerable. 

A machine of considerably different type is shown in Fig. 67. It is a hand-operated 
welder specially designed for light work on steel strips, small rings, perambulator tyres, 


cycle parts and such-like small articles. In the illustration the casing of the lower 
part has been removed in order to show the construction of the transformer. The 
secondary winding consists of laminated copper strip, which encloses the primary coil 
and embraces one limb of a rectangular iron core. The endd of the secondary winding 
are bolted to fixed platens which carry the clamps and pressure device. The elsimps, 
which are of the vertical type, are operated by cam levers, and are so arranged as to 
enable the work to be inserted from the front, and to be gripped by a single movement 
of the cam lever. The pressure device consists of a toggle and lever, and there is an 
automatic switch, which, however, is not shown in the engraving, by means of which 
the primary current can be cut off as soon as the sliding clamp reaches an adjustable 
stop. This machine is, we are informed, capable of welding steel strip as thin as 
No. 22 S.W.G., two inches wide, with uniform and reliable results when operated 
by a lad. 

An entirely automatic welder is shown in Fig. 68. It is intended for welding hoops, 
rectangular frames, buckles and rings at a speed which may amount to as manj'^ as 
600 welds per hour. The transformer and the electric parts of the machine are of 
standard type, but the clamps are mechanicaUy operated by a system of cams from 
a main shaft which is driven by a clutch pulley controlled by a treadle attached 
by chain to the rod suspended from the pulley at the right-hand side of the machine. 
The machine is belt-driven, and as soon as it is started up the right-hand platen moves 
away from the left, with the result that the space between the arms of the clamping 
device is increased and the clamping jaws raised. As soon as this occurs the operator 
puts the pieces to be welded on the electrodes, so that they abut against each other 
evenly with the joint in the centre. The clamping jaws then descend and securely 
hold the pieces in place. The primary switch then closes and the current flows. 
As the metal softens the pressure of a spring causes the right-hand platen to move 
towards the left, with the result that the metal is upset at the points, and the weld 
made. The steel clamping jaws are then raised, the right-hand clamp then again 
moves away from its fellow, the operator removes the welded piece, and the cycle of 
operations is again repeated. 

This machine is of 6 kilowatts capacity, and is intended for welding stock up to 
I square inch cross section, at the rate, as already said, of 600 welds per hour, the 
actual time required for heating each weld being, we gather, about three seconds. In 
both this machine and in that shown in Fig. 57 there is a circulating water-cooling 

SoMB Large American Spot Welders 

Some exceptionally large resistance welding machines, which give some idea of the 
far-reaching possibilities of this method of welding, have recently been constructed 
by the General Electric Company of America. First of all a fixed experimental 
machine was built which had a welding current capacity of no less than 100,000 
amperes, and a pressure capacity of 76,000 lb. As a matter of fact the maximum 
current and pressure at which the machine had been used up to the time at which 
the memorandum from which we are obtaining our information, and which appeared 
in the issue for December, 1918, of the General Electric Review, was written, were 72,000 
amperes and 30,000 lb. respectively. With them three thicknesses of 1 in. plate had 
been spot welded. Working on the data obtained with that machine the company 
set about the designing of special portable appliances for use in the fabrication of 
structural parts, particularly intended for ship work. An investigation of the subject 
showed that four-fifths of the work of this character could be carried out by a machine 








having a reach of 12 in., and that a machine with a reach of 27 in. would include the 
other fifth. Since it was calculated that both the weight of the machine and the 
kilovolt-amperes required for its operation would be about 33 per cent, greater for 
the 27 in. reach machine than for that having the 12 in. reach, it was decided to build 
two machines only that with the longer reach. The standard pneiunatic 

riveter w&a U lei, w far as the general construction of the machines was 

concerned, th pressure being applied by means of an air cylinder and a 

lever system. ilso provided so that the machines could be picked up by a 

crane and hel where required. An engraving showing the larger of the 

two machines ;. 59, which is reproduced from the Review mentioned above. 

The maxi: ; current available in these^ machines, n-ith a steel plate 

enclosed to tl; f the gap, is about 37,&00 amperes. I'he current applied to 

the primary winding had a periodicity of 60 and a volti^e of 534. Reduced voltages 
to give pmaller currents in six progressive steps 
— the lowest being 267 volts — were furnished by 
means of separate regulating transformers. This 
. arrangement was provided to meet the possible 
requirements for a considerable range of thickness 
of work, as well as for experimental purposes. As 
a matter of fact, however, it was subsequently 
ascertained that the machines would operate satis- 
factorily, on work of thicknesses extending over 
the range on which they are likely to be used, when 
connected directly to a 440-volt, 60-cycIe current 
with no regulating transformers. Two plates are 
with these machines welded togetherin spots from 
1 in. to Ij in. in diameter in from twelve to fifteen 
seconds. Thicker plates, of course, require more 

T«.._„.. , ^ time, and thinner plates less time. 

Ft<i. 09. The welding current under these conditions is 

American Spot Welder tor Ship Work. about 31,000 amperes, the corresponding primary 
current for the 12 in. mRchinc being fibout 600 
amperes, and for the 27 in. machine about 800 amperes, the voltage in each case 
being 440. The expenditure of energy was therefore 264 and 352 kilovolt-amperes 
in the two machines respectively. The secondarj- voltages with these iigures work 
out at about 8 J for the smaller machine and 11 J for the larger. 

The maximum mechanical pressure for which ther.e machines were designed is 
25,000 lb. It is obtained by means of an 8 in. cylinder working with air at 1001b. 
per square inch pressure, and a lever arm with a ratio of 5 to 1. Alteration in the 
pressure apphed is obtained by varying the air pressure, a reducing valve being 
provided fo,r the purpose. A gauge pressure of 70 lb., giving a pressure on the work of 
17,500 lb., has been found to give good results with J in. plates. 

The conductor for the primary windings of these transformers m | in. by ^ in. 
annealed copper tubing, insulated with asbestos tape, wound directly on an insulated 
core, alternate layers of sheet asbestos and mica pads being used between layers of the 
primary winding, between primary and secondary, and between primary and the core. 
There are four layers of thirteen turns each in the 12 in. machine, and three layers of 
thirteen turns each in the 27 in. machine. The U-shaped single-turn secondaries, 
which were slipped over the outside of the primary windings, were constructed of two 
copper plates, each | in. thick and 6| in. wide, assembled one inside the other, J in. 


apart. Xarrow Btrips of copper were inserted between the plates along the edges, and 
were brazed in poeition bo as to make a water-tight passage for the circutatioa of the 
cooling water. At 31,000 amperes the current densities in the secondaries ia about 
6200 amperes per square inch, the corresponding densities in the primary windings 
being about 7000 amperes per square inch in the 1 2 in. machine and 9000 in the 27 in. 

Spacing blocks of a compound of asbestos and Portland cement were used at the 
onds of the core and at the ends of the winding layers, and the complete transformer, 
after assembly, was impregnated with " bakelite." The result, we gather, was a . 
solid mechanical unit which is not injured by heat as long as 150 deg. Cent, is not 
exceeded. Several welds can, we are informed, be made without turning on the cooling 
water before that temperature is reached. The cooling water wag given two paths, 
one being through the primary winding, and the other through the secondaries and 
the electrodes in Series. 

When it was found that these machines worked satisfactorily it was decided to go 

Fio, 00. — Lai^ American Duplex Spot .Welder. 

a step further, and to build a larger machine, intended specially for the application 
of electric welding as a substitute for riveting on parts of ships hulls composed of 
large-sized plates, which may be "fabricated " before they are assembled in the 
ship. It was decided to give the machine a 6 ft. gap, and to make it capable of welding 
together two plates } in. thick in two spots at the same time. It was realised that 
it would be useless to attempt to make the machine even semi-portable and the form 
it was given is shown in Fig, 60. With the welding circuit enclosing a 6 ft. gap, and 
carrying the very heavy current necessary to weld | in, plates, the kilovolt-ampdrage 
would have been very high. It was proposed, therefore, as a means of greatly reducing 
the energy consumption, and at the same time of doubling the amount of work turned 
out, to use two transformers with two sets of electrodes, and to arrange them as shown 
diagrammatically in Fig, 61. By this arrangement it was made possible to weld two 
spots at the same time. The transformers are mount«d in the frame of the machine 
as neai- to the welding electrodes as possible, so as to obtain the minimum reactance 
in the welding circuit. The polarity of the electrodes on one side of the work was made 
the reverse of that on the opposed electrodes, thus giving a series aiTai^ement of the 
transformer secondaries, the pathway of the current being indicated by the dotted 




hsMsH, It will be gathered that the cmrent from each tmisform«' Oaws through both 
of the KpotM being welded. The bottom eleetrodee are j^tatiooaiT, while the top elec> 
trode*$ are capable of independent vertical motion, so as to engage the work. Previous 
teiftn with the original experimental machine had shown that soceessfoDy to weld 
two KpotK at the iiame time with such an arrangement as we have described, it was 
ntcei^hary that the pressure should be indepoidently applied. Otherwiae, because of 
inequalities in the thieknetss of the work, or because of wear and tear of the dectrodes, 
the preM»ure might be mu ?h greater on one of the spoti> than on the otlier, which would 
naturally result in unequal heating, since the resistance to the flow of the electric 
current and the consequent heating effect are less on the spot with the greater pressure. 
The two top electrodes were therefore mounted on separate plungers operated by 
separate pistons through independent levers. 

The pressures obtained in this duplex machine with an air pressure of 100 lb. per 

square inch are 30,000 lb. on each spot, 
or a total pressure of 60,000 lb. on the 
work. The two air cvlinders are 
mounted on a cast iron bfed-plate in 
the back part of the machine. The 
levers connecting the pistons to the 
electrode plungers, which are 7 ft. in 
length, were made of cast steel in order 
to obtain the necessary strength. 

The maximum welding current for 
which this enormous machine was de- 
signed is 50,000 amperes, this secondary 
current being obtained with 60 -cycle 
current at a voltage of 500 applied to 
the primary winding. With this current 
in the secondaries the current in the 
primary windings was 1800. The 
primaries of the two transformers are 
arranged in series and the kilovolt- 
ampdrage in each transformer is 450, 
or a total of 900 kilovolt-amperes. The 
current densities in the primaries and 
the secondaries of these transformers 
was no less than 9000 and 7000 amperes 
per square inch respectively. The winding consisted of two layers of fourteen turns 
each for each of the primaries, and a single turn for each of the secondaries, one 
layer for the primary being arranged inside and the other outside the single turn 
secondary winding. ThcHe tranrformers, which, exclusive of terminals, only measure 
11 in. by 16 in. by 18 in. over all, are thought to be the smallest ever built for so 
large a rating. 

It can bo well understood that when dealing with such heavy currents the question 
of the electrodes becomes of great importance. In the portable machines described 
above the current density in them is given as being about 60,000 ampdres per square 
inch. It has to be remembered, too, that when heated by the passage of this current 
the tips of the electrodes are subjected to a pressure which may amount to as much 
as from 16,000 lb. to 20,000 lb. per square inch, and that copper— the best available 
material for the purpose— softens at a temperature considerably below the welding 

To tmrmltml 
of primary 

of uppm 

Flo. 61. 
Electrical Arrangementfl of Duplex Welder. 



temperature of steel. The electrodes, which have tips in the form of very flat truncated 
coned, are given such shape as is calculated to afford as free conduction as possible of 
the current to, and of the heat away from, their tips. Moreover, they are adjacent 
to large masses of metal of high electrical and heat conductivity, and their bodies are 
internaUy water-cooled by a stream of water flowing continually through. Still, in 
spite of every precaution, which included the clearing away by meaniS of a sand blast 
of rust and scale from the stirfaces of the plates at the points of impingement of the 
electrodes, deformation of the tips of the electrode occurred, thus increasing the area 
of contact with the work and reducing both the current and the pressure densities 
below the values needed for welding. To overcome this difficulty thin copper caps — 
^^ in. thick — ^fitting over the tips of the electrodes have been tried, and as many as 
160 welds have been made with a single cap before it required to be renewed. Another 
method tried was to have renewable tips to the electrodes, and it has been proposed 
to employ electrodes which combine the features of the removable tip and the cap, 
but the last-named had not been tried at the date of the Review to which we are 

As far as we are aware, the machines described in the foregoing are the largest 
and most powerful which have, up to now, been constructed. 



In an earlier chapter we mentioned that the Steel Barrel Company, Limited, of Ux- 
bridge, in addition to its extensive use of the carbon arc, also employed the resistance 
welding process. It does so for the production of sheet-steel drums such as are used 
for containing petrol, petroleum and other such oils and liquids, and it finds it most 
satisfactory for the purpose. We ourselves have seen a five-gallon drum made by this 
method at Uxbridge which, when filled with petrol, and weighing in all about 49 lb., 
had undergone the drastic test of being dropped no less than seven times from different 
heights — nearly 20 ft. was, we believe, the highest — and it did not show the least 
sign of leakage, though it was severely battered about and distorted. We hiyve on 
several occasions watched the various processes of the manufacture of these drums 
and we shall now proceed to describe them. 

It may be explained at the commencement that some years before the outbreak of 
hostilities, Mr. T. T. Heaton, the general manager of the company, when on a visit to 
Germany saw in operation a series of machines which he thought could be adapted to 
perform the kind of work which he had in view. A set of machines was, therefore, ordered 
and eventually delivered. As originally built, however, they failed to do what was 
required of them, and had to be considerably modified, with the result that in their 
present forms they certainly operate extremely well, and with great speed. The size 
of drum which we have seen made was that intended to hold five gallons, but the same 
machines are equally well fitted for making larger or smaller vessels, the largest or 
smallest for which they are suited being, we understand, 2 and 15 gallons respectively. 
For making the drums four distinct types of machines are employed, and one type is 
represented by two slightly different machines, so that it may be said for the welding 
alone five different machines are required. We shall refer to them in the order in which 
they are used. 

The drums are made of mild sheet steel of Nos. 24 to 18 B.G. thickness. The 
sheets are first of all cut to correct size, and accurately squared. They are then curved 
into cylinders, by passing them through rolls in the usual manner, the pressure being 
so adjusted that only one passage through the rolls is necessary in order to obtain 
the desired curvature. To obtain really satisfactory welds by the processes employed 
it is necessary that the sheets should not only be free from scale and rust, but that 
aU oil, grease and dirt should be removed from their surfaced. To effect this the 
cylinders, when rolled, are immersed for a short time in a bath containing a mixture of 
acids diluted with water. When they have remained for a sufficient time in the bath 
they are taken out, washed in several changes of water and dried. In this connection 
we may mention that, as the washing water has, after use, to be discharged into a 
canal which passes by the works, its acid contents have to be neiitraUsed. To do this 
it is passed through a channel which is charged from time to time with the spent 
material — calcium hydrate — taken from the acetylene generators. We understand 
that this employment of an otherwise useless waste product has given quite satisfactory 


The cleaned and dried cylinders are then ready for the welding prooes^s, the firBt 
of which is to give the necessary overlap to the edges of the longitudinal seam^ and 
to secure these edges in the correct position for the welding of the seam. The machine 
employed for the purpose is a pedal-worked spot welder, the lower electrode of which 
is fixed, while the upper is reciprocated vertically by the pedal. The machine is mani- 
pulated by two boySj one of whom only assists in holding the cylinder in the required 
position. The correct overlap is obtained by causing the longitudinal edges of the 
rolled cylinder to enter two slots arranged one on one side, and the other on the other 
of a horizontal bar, the two slots being at slightly different levels, and being given 
such depths that when the edges of the curved cylinder plates are pressed as far as 
they can go into them by the two boys, one pressing on one side and one on the other, 
the exactly correct overlap for the joint is obtained. The machine operator then, 
after making certain that the ends of the two edges, which, we may explain, are aUowed 
to project some little way from the end of the slotted rod so that they can be readily 
pressed together, coincide, or, in other words, that the circumferential edge lies in 
one plane, adjusts the extreme end of the seam immediately beneath the vertical 
electrode of the machine and presses the pedal. The result is the formation of a minute 
spot weld, which effectually holds the two edges together with the correct overlap. 
The amount of overlap being given when we saw the machines at work was about 
three-eighths of an inch, and the diameter of the spot weld about one-eighth of an 
inch. Still keeping the edges of the cylinder in the slots in the bar, the cylinder is 
moved horizontally for a short distance, and another spot weld made. This process 
is repeated several times, until only the length of seam at the end remote from the 
first spot weld remains. The cyUnder is then removed from the machine, the unsecured 
end of the seam is adjusted in position on the lower electrode, and the upper electrode 
brought down by the pedal so as to make the final spot weld. The whole process, 
which may from the description appear to take a long time, is actually carried out in 
a few seconds, and results in the formation of an accurate cylinder, the edges of the 
longitudinal seam of which are held securely in place for the next operation, which 
is that of welding the seam. 

For the welding of the seam the cylinder is placed in the horizontally reciprocating 
frame of a machine, which is furnished with revolving wheel electrodes arranged, 
vertically, one above the other. When in position in the frame with the seam of the 
cylinder coming immediately between the two wheels, the movement of a horizontal 
. lever causes the top wheel to be moved downwards, so that the seam is pressed between 
the two wheels, while simultaneously the horizontal frame, and with it the cylinder, 
is made to move slowly in such a way that the wheels, or rollers, as they revolve with 
the seam between them follow the latter accurately and weld the two edges securely 
together. As in the case of the fir^t machine the whole length of seam is not welded 
from one end. The welding of the seam is begun three or four inches from one end, 
and the weld taken from that point to the further end of the seam. The cylinder is 
then taken out of the frame, reversed, and put into the frame again. The remainder 
of the seam is then welded in the same manner. The whole process, including the 
reversmg of the cylinder, takes only about ten seconds. By reason of the fact that the 
cylinder is held in and moves with the frame, the weld is in an exactly straight line 
and looks extremely neat and busine^sUke. 

The next process is the fitting and welding in of the top end of the drum, that is 
to say, that end which carries the bunghole. The ends, whether that at the top or that 
at the bottom, are formed of circular steel plates of Nos. 24 to 18 B.G. thickness, 
dished so as to have a rim standing up at right angles to a depth of about 1 in. The 


ends when dished are of such a size that they are an easy driving fit into the end of the 
cylinder. The hole to take the bung ring is, of course, punched before the end is 
fitted into the cylinder. After the end has been placed in position, an operation which 
is effected simply by pushing, or, if necessary, light tapping, the operator satisfies 
himself that it is neither too far in or too far out, and then places the cylinder horizon- 
tally on a frame in a welding machine, the arrangement being such that the cylinder 
can revolve on its axis. The welding machine is of the wheel tj^e, the two wheels 
being placed vertically one above the other. The top wheel is revolved by power, 
while the lower wheel is free to revolve on its spindle. All being in readiness the 
machine attendant depresses a lever which raises the bottom wheel so that the rim 
of the cylinder end is gripped between the two wheels, and welding commences. As 
the upper wheel is revolved mechanically the drum cylinder is revolved with it, and 
gradually the whole seam is welded right round its periphery. . All that the attendant 
has to do is to see that the rim is kept continually between the two wheels, this being 
done by guidance with the hand, which ensures that the end of the cylinder revolves 
in one plane. This operation, like those which have gone before it, is also effected in 
but a few seconds of time. 

The next operation is the welding on of the bung socket, or boss, which may be of 
two types, i.e. that for use with a cork bung, and that for a screw plug. In both cases 
the welding process is the same. The cylinder is threaded over an upright, the top of 
which is provided with a spindle which fits into the hole punched in the end piece, 
and is provided with a shoulder on which the metal of the end piece can rest, and 
which forms the lower electrode and the anvil for the welding operation. With this 
arrangement, as will be understood, the cylinder can be revolved vertically round the 
bung hole as a centre. When the cylinder is in position the bung socket is threaded 
over the projecting top of the spindle above mentioned, and the machine put in motion. 
The effect is to cause an upper electrode to be mechanically reciprocated vertically 
up and down, and at the lowest point of each stroke to press the metal of the cyhnder 
top and the flange of the bung socket together, and form a spot weld. Between each 
stroke of the machine the cylinder is revolved through a small angle, the effect being 
that a series of spot welds, which just overlap one another, is formed all round the 
periphery of the flange of the bung ring, and when the cylinder has been revolved 
through a full circle the welding is completed. The result is neat and satisfactory in 
every way, and the whole operation is very quickly effected. 

Next comes the welding on of the handle, which is simply a strip of sheet steel 
bent through four right angles in the following manner ^l l^. The 

machine used for the welding is of a similar type to that used for welding in the bung 
socket, but of a slightly different construction. There is, however, a vertical lower 
electrode which also acts as an anvil, and a power-worked vertically reciprocating 
upper electrode. The handle is placed in position and some five spot welds quickly 
made on each of its ends where they come in contact with the metal of the drum top. 
These welds, though they look quite small, are sufficient to weld on the handle so 
securely that a very considerable pull would be required to tear the handle from the 
drum top. 

Finally, as far as the welding processes are concerned, the bottom piece, which 
exactly resembles the top piece, saving that it has no bung hole, is fitted in and welded 
in place, the machine employed being identical with that used for welding in the tops. 

The finished drum is then taken to a grindstone, by means of which the rough 
edges of the ends are smoothed, and it is then ready for testing and painting. Testing 
is effected by inserting a closely fitting nozzle into the bung hole, applying internal 


air pressure, and plunging the drum in water. Out of a large number which we saw 
teg^ted we saw no sign of air leakage. Drums of this tjrpe are turned out in large 
numbers and with wonderful rapidity. There is not, we should say, a single one of 
the welding operations which takes a quarter of a minute, the shipping and unshipping 
of the cylinder in the machine being included in that period. 

Having described the welding processes, we can say a few words regarding the 
machines. They all, of course, work with alternating current. The company does 
not generate alternating current nor does it convert its own direct current into alterna- 
ing. It ^o happens that a supply from a local company is available. The voltage 
at which it is received on the works is 2800, and the periodicity is 60. Near the point 
at which the current enters the works the pressure is transformed down to 220 volts, 
at which pressure it is distributed to the resistance welding machines. The secondary 
voltage of the latter, that is to say the difference of potential between the elegtrodes 
applied to the work, is very low. In no case id it, we understand, more than 1 volt, and 
in one machine, at any rate, it is as low as half a volt. In this, connection it has to be 
remembered, of course,, that the material being dealt with is very Ught, only Nos. 24 
to 18 B.G. 

Of the machine used for spot welding the longitudinal seams there is not much 
to say beyond that which has already been said. The motion involved is simply the 
reciprocation of the upper electrode, which lA effected by pedal against a spring. Just 
after the final pressure is put on, the current in the primary circuit is automatically 
cut off. Both upper and lower electrodes are cooled by water circulation. 

The longitudinal seam weldmg machine, on the contrary, has many points of 
interest. Its two electrode wheels, the spindles of which are carried in bearings, are 
mounted at the ends of two long arms which form the two terminals of the secondary 
of the transformer. They are both of them free to revolve and are cooled by a large 
portion of their surfaces being in rubbing contact with cheeks which are provided 
with a water circulation system. Rubbing contact is depended upon for the conduc- 
tion of electric current. The mechanical arrangements are worthy of special mention. 
It will be remembered that the movement of a horizontal lever both brings the elec- 
trode wheels nearer together so* that they press on both sides of the work, and also 
puts the cylinder-carrying frame in motion. The action is, briefly, as follows : The 
upper arm, which carries the upper electrode wheel, is hinged near its rear end, that 
is the end away from the wheel, and the rear end itself is pin-connected to a system 
of levers which are actuated by the horizontal lever. The motion of the latter im- 
parts just sufficient motion to the arm carrying the wheel to enable the latter to bear 
with the desired pressure on the work. Connected to the horizontal lever is also 
the actuating member of a clutch which can put a train of revolving gearing into and 
out of mesh with a toothed wheel on a shaft carrying a pinion which is in mesh with a 
toothed rack attached to the horizontally reciprocating frame which carricEl the 
cylinder. There is an arrangement by which the frame when it has completed its 
forward movement is quickly returned to its starting-point. The power to work this 
machine, and the others that require it, is obtained by belting from a countershaft. 
It will be understood that in this case the current is kept on during the whole of the 
welding operation, and that as the cylinder is forced between them the electrode wheels 
revolve so that no one point in their peripheries is kept long in contact with the heated 
metal which is being welded. The weld is continuously being consolidated by the 
pressure between the electrode wheels. 

The top and end welding machine is, like that which has just been described, 
exceedingly ingenious. The upper wheel, as we have said, is kept continually revolving, 


the lower wheel being free on its shaft. The latter, of course, revolves when work is 
being done. The lower carriage wheel is capable of vertical motion by means of a 
system of levers which produce a certain pressure on the work when it is gripped 
between the wheels. In order, however, to allow for irregularities in the thickness 
of the material passed between the wheels, as, for example, the overlap of the longi- 
tudinal joint, the carriage of the lower wheel is furnished with a strong coiled spring 
which allows of just sufficient '* give " to prevent injury to the machine or undue 
pressure on the work. In this machine also the current is kept on during the whole 
period of welding. Cooling of the electrode wheels is effected in a manner very similar 
to that employed for the longitudinal seam machine. The electrode wheels in both 
machines are of pure copper. It has been found impracticable to employ an alloyed 
material, since the alloy is eventually volatilised leaving the copper in a spongy state. 
During use the contact surfaces of the wheels become coated with a more or less non- 
conducting surface, formed, possibly, by oxide from the steel plates. This is removed 
at intervals by light scraping. After continuous work for perhaps a fortnight, the 
wheels may get slightly out of shape. If this be the case they are removed from the 
machine and a very light cut taken from the peripheries in a lathe. 

The reciprocating spot welding machines employed for affixing the bung socket 
and handle do not call for any extensive description. The reciprocating motion is 
mechanically operated, the upper electrode in the case of the former machine making 
strokes at the rate of about one per second, or even less, which appears to give ample 
time for the operator to revolve the drum through the requisite angle between each 
stroke. Each and all the machines work smoothly and well, and produce excellent 
results. ' Taken altogether the whole system turns out drums of first-rate quality 
in a wonderfully expeditious manner. 



A FIRM which, though it has not been in existence for many years, has nevertheless 
developed some excellent welding machines of various types is the A 1 Manufacturing 
Company, of Industry Works, Sunbridge Road, Bradford. It confines its activities 
entirely to resistance welding and to the production of machines and accessories for 
welding on that principle. It contends, and with some show of right, that resistance 
welding and its possibilities have been to a large extent outshone and overlooked by 
reason of the extensive publicity which has recently been given to arc and acetylene 
welding. It points out that neither of the two latter processes lends itself to any- 
thing like the same extent to repetition work as does resistance welding, and further, 
that whereas both of them require a no inconsiderable degree of skill for the production 
of good Eiound welds, there are many directions in which quite unskilled labour can 
achieve excellent results with resistance welders without the necessity for any special 
training. The comfnent we would make concerning these contentions is that it was 
largely with the idea of correcting wrong impressions regarding electric welding as 
a whole that the publication of this series of articles was undertaken. Electric welding 
in various branches has been practised for a number of years now, but it must be ad- 
mitted that the multifarious nature of its applications has not, until comparatively 
recently, been recognised by the large majority of our manufacturers. There is little 
doubt, however, that the future will see its adoption far more extended than it is even 
now, and it will be found that there will be ample scope for both the resistance and 
arc systems ; for each will be discovered spheres in which it operates with best effect. 
This by the way. 

Let us now turn to the firm's machines. We have before us a list which shows that 
it has evolved as standard patterns eight spot welders numbered to 7, three seam 
welders, Nos. S3, S4 and S5 ; and eight butt welders, numbered BOO to B8. The 
rated continuous working capacities of the spot welders vary from -j-^ in, added thick- 
ness in the smallest mp«chine to | in. added thickness in the largest machine, the 
material dealt with being mild steel and iron. For intermittent working the thick- 
nesses may be increased to 3^ in. and J in. respectively. *' Added thickness," it need 
hardly be explained, is the total thickness of the pieces of metal that are to be welded 
together. The six smaller machines are also stated to be capable of welding brass and 
aluminium, the thicknesses dealt with var3dng from ^^ in. with machine No. to ^^ in. 
with machine No. 5. The smallest seam welder is designed to take 3% in. added thick- 
ness, and the largest ^ in. with mild steel or iron, the thicknesses with brass being 
just half those figures. The diameter of material which the smallest butt welder — 
No. Boo — is intended to take is ^^ in., while the largest machine — B8 — is designed 
to operate with material up to 2^ in. in diameter, or, say, 5 square inches in cross 
section — the figures given relating to ferrous metals. Copper and brass can be welded 
in the five smaller machines, the diameters varying from .02 to .0625 in the smallest 
machine to | in. for copper and | in. for brass in machine No. B5. In addition to the 
foregoing, there are four sizes of seam welding attachments and three butt welding 



attachments which can be applied to different sizen of the finn's spot welders, so that 
the latter can be temporarily converted for the purpoxes named. The four eeatm 
welding attachments can deal with added thicknesBes of -^^ in., ^ in., ^ in. and ^\ in. 
respectively, while the three butt welding attachments can take diameters of from 
^ in. up to ^ in., | in. and J in. respectively, the sizes given referring to mild steel or 
iron in each case. 

The foregoing are standard machines and appliances, but they do not by any me&ns 
constitute the whole of the firm's output, for a not inconsiderable proportion of its 
business consists in the design and manu- 
facture of machines specially worked 
out and constructed to meet the peculiar 
requirements of clients. 

In principle, of course, the "A I" 
machines are identical with other resist- 
ance welderH. They a31 work with 
alternating current, and each contains 
a static transformer, the primary wind- 
ing of which has many turns, while the 
secondary, in general, consists of one 
or two turns only. It is in their details 
that these machines are differentiated 
from the products of other makers. 

As typical of the company's spot 
welders, we give in Fig. 62 a view of 
what is known as the No. 2 spot welder. 
It is a machine which is specially in- 
tended for the production of hollow- 
ware and general light work with iron 
or mild steel \ip to J in. added thickness 
for continuouR working, and up to ^ in. 
added thickness occasionally. As a 
maximum it requires 6 kilowatts of 
energy to operate it. The standard 
clearance of the stakes or arms — that is 
to say, the distance clear from the 
electrodes to the back of the arms, is 
9 in., and the top arm is so designed 
that it can be extended to 16 in. in 
length as required. With arms of 
standard length the net weight of the 
machine is 462 lb. Stakes of greater length, so as to give clearances of 18 in., 24 in., 
or 36 in., can, hoM'ever, be fitted if required, and the weight with the longest stakes 
becomes 501 lb. The top arm or stake is hinged at the same level as the top of the 
lower electrode, with the object of ensuring a straight pressure on the weld and none of 
the current passes through the hinge, so that the latter docs not become heated. The 
top and the bottom arms are both fitted into round socketM, so as to allow them to be 
twisted to the right or to the left, so that awkward positions of the weU may be reached , 
while the lower electrode can also be moved up and down in the slots formed in its 
carrier and the latter can, itself, be moved from side to side. In the standard machine 
the top arm is solid, and the electrode it carries is water-cooled, as is also the lower 

Fio. 62.— "A I" Siwt Welder. 


electrode. The upper electrode, as will be observed in the engraving, is of considerable 
length, so as to allow of a long range of adjustment of the lower electrode arm. The 
upper electrode is pierced at its lower end with a hole drilled to a Morse taper so as to 
take the shank of a tip piece, which can hence be replaced when it is worn or deformed. 
The lower electrode tips are also renewable, but the tips are screwed in and not tapered. 
The renewable tips are conical in shape, and 
are made flat at the extreme tip. It will be 
noticed that the hinged upper electrode arm 
is fitted with an adjustable balance weight, 
BO that different lengths of overhang of the 
arm can be allowed for. By it the force 
necessary to operate the machine may be 
reduced to a minioium. 

It is said that on small articles of, say, 
30 B.W.G. mild steel, that can be quickly 
handled, from 40 to 50 welds per minute 
can easily be made with this machine. The 
standard winding is for 50-period single- 
phase alternating current at 200 volts, but, 
of course, other windings can be supplied if 
required, though for them we understand 
that an additional charge is made. The 
slotted box seen at the side of the machine 
is a change-over switch by which nine 
separate tappings on the primary winding, 
whichwill give nine strengths of current.may 
be obtained in the secondary winding. The 
speed of working can -therefore be varied 
over B. wide range. No. 1 is the fastest and 
No. 9 the slowest, and there is an " off " 
position by means of which the winding is 
cut right out of circuit. In some of the 
firm's machines this switch is replaced by 
a series of plugs, there being a fast (F) plug 
and a slow (S) plug and tour further plugs 
available for use with either the S or the F 
plugs, the combination giving eight working 
speeds. A thimble is provided on the 
pedestal into which can be sweated an 
earthing uii'e, so as to remove the possibility 
of the operator receiving a shock should 
the frame of the machine accidentally be 
made " live." P,^,^ 63.-" Al " Automatic Sttitcli. 

The machine is worked by pedal, which 
can be swivelled so as to be arranged in the position most comfortable for the operator, 
and this pedal actuates a vertical rod fitted at the rear of the machine. The top of 
the rod passes through a hole in the end of the upper electrode rocking arm. 

The upper part of the rod is threaded through a spring, which, when the pedal is 
operated, is compressed and causes the upper electrode to be pressed down on to the 
work and eventually to consolidate the weld. The initial compression of the spring can 


^^ r^/jkUtd t>T ■:«aci'« of a c-^t. «o ai^ to ^-'^v f^.r ?h^ dif «VK:t pffV!<««rM required when 
w^ifiinf iiHltrttA tfcir^krj^^*^^^ of mtrtiL TLene- i« a f^^riber «pnzi2 wlurh aru as a baffer 
ar>i ta.ka» T;p th^ «Lo^k viu!tk tbe* pre^^Tire' c*ei tbe* pedu Lf «addciilT refirrcd. 

To eomfrl^^ tr^ e<(r;i|Miiei;t «( iht ma^Liue tr^nv t^ aa aaiomatie Fvitch which is 
'.f ^rar^ by the f/^d^-wr^td rertical red. aod vL>h pcniiit:^ norent lo enter the 
f,r\gr.s^ry wir^iir^ vh^^ the vcrk L^ lightly zripf^d between the <ffcitmd e» and eiits it off 
z^ 'ocm a* w»4riir^ heat i^ reaeb^d. In operacion the work i^ H^htly gi iyped between 
^h^ '-i^^^rod^^ by the depre^«ion of the pedal cao^in^ the actnadng spring to be com* 
pr»^<^. A further depre^ion of the pedal — ^with a re«nhing additional com p r e saon 
of tfje •pfiTig — faiLRe'i the «witeh tirj^inz the primary eirrait to be operated, while a 
•t.J frrrther depre^^ion of the pedal trrp^ and open« the switch and ftiD further com- 
pr^.#-« the •prin^. with the result that the weld l« firmly consolidated. 

The maker* claim for their «pot weUer?. of which the foregoing may be eonodered a 
t;« pica! example, that tb^ are practically fool-|M^>f. and that they can be operated by 
a ^^r•\ girl or other ari.«killed labr>iirer. and that about an hour's practice is all that is 
rei^uircd to enable a per-on of average ability to make thoroiighly satL^actory welds 
cviTftinuofi^ly on a large variety of repetition w<H'k. 

The arrangement of the automatic trip x^witch is shown in Fig. 63. The vertical 
a^rtuating bar l'^ !»hown at A. Mounted to slide up and down on it is a tripper carrier B, 
whieh can be clamped in any desired position, so as to allow for different thicknesses of 
met ^1 being welded and different travels of the electrodes by means of the small hand 
!e%'er B^. The trip plnnger C slides in a hole <m the carrier B. and is continualh- being 
pre»«ed forward by the action of the spring D. but is prev«ited from being pushed out 
iA the hole by reason of the set s^crew C ^. Wlien the actuating bar A is preyed upwards 
by the depression of the pedal by the operator, it takes the carrin- along with it and the 
trip plunger C engages with the catch piece E. which is clamped in a hcSe formed in the 
k/wer p^.rtion of the hinged switch arm F. which is pivoted at G. The catch piece E 
can be given any desired amount of projection, and is firmly clamped in position by 
roeanM of the bolt E*. The effect of continued depression of the pedal is that the 
^witch arm, which isi normally held in the open positi<m by the action of gravity, is 
rcnrolved ahout itH pivot and the contact pieces J J' are brought into contact with 
the contact pieces H H* respectively, and the primary circuit of the transformer in the 
machine is closed. Current then, el course, flows through the secondary winding and 
through the metal being welded, since arrangements are made so that the switch is not 
clr^scd until the work is firmly gripped between the electrodes. Still further depression 
€ji the pedal unll cause the springs K and K^. which encircle the cont<act rods H and H^, 
to be com premised to such an extent that the trip plunger C becomes disengaged from 
and can slide pant the catch piece E, with the result that the switch flies open and the 
circuit is broken. The period of time during which current is permitted to flow may 
he regulated by the various adjustments provided. The material of which the contact 
nxls of the switch are made is a yellow metal termed by the makers " non-arcing 
mortal/' Its composition was not disclosed to us. but we understand that it has been 
found to be quite ftucce»ij<ful, no trouble of any kind having been experienced with it. 

The other npot welders made vary in certain details from the machine described in 
tlHT foregoing, but they all operate on the same fundamental principle. 

A typical example of the firm's smaller sized butt welders — ^which represents the 
nriachine known as welder B3 — is shown in the drawing Fig. 64. In it the two pieces 
of material to be welded are clamped in the gripping jaws A and B by means of the cam 
handU'K A* and B* rcHpectively. The jaw B can be reciprocated horizontally by means 
of the hand return lever D, which is pivoted to a sliding carriage C. The initial position 


of the lever D can be varied to permit of different distances apart ol the jaws by chang- 
ing the length of the hnk D', in which several holes ore formed for the purpose. The 
pressure for upsetting the weld is pro- 
vided by a spring, wWch is contained in 
the sliding carriage C. The initial com- 
pression of the spring can be adjusted 
by means of the small star wheel E, and 
the screw of the spring spindle can be 
locked by the star wheel F, It will be 
understood that the first motion to the 
left of the hand return lever D brings 
the two pieces of work into contact. 
Further motion of the lever puts more 
compression on the spring, and henco 
increases the pressure between the pieces 
being welded. 

The electrical arrangements are 
somewhat different from those of the 
spot welding machine. There is, in 
the first place, provision for only 
five working speeds. Then there are 
two switches — a push switch and an 
automatic switch or circuit breaker. 
The normal- position of the latter is 
with the circuit closed so that if the 
push switch is in, current begins to flow 
as soon as the two pieces being welded 
come into contact. The movement of 
the hand lever and of the sUding carriage 
C has the effect of bringing the tripping 
piece G into contact with the switch 
trip tool, and when a predetermined 
point has been reached the piece G 
forces open the switch and breaks the 
circuit, the arrangements being such 
that the switch is thereafter kept open 
while the lever is held to the left, since 
the tripping piece impinges on the fiat 
under surface of the switch trip tool. 
When the upsetting of the weld ia com- 
pleted the spring push switch is released 
by removing the hand from it. The 
welded work can then be taken from the 
jaws, and the lever D moved to its 
starting position. 

In oidinary working the amount of 
travel of the sUding carriage C is con- 
trolled by the adjustable slide stop E, which ensures that when there is no work 
in the machine the jaws themselves can never be brought into contact with one 
another. For use when it is considered necessary to anneal the welds — as, for 

Fio. 61.-"A1" Butt Welder. 


in>laooe. vlien dealing with high-carbon ctnJ — aoothtr adjnctaiilr elide itop F, 
which i« embodied in the name eaatii^ a« the •-lide stop E. it nraog into the positiim 
ot-tiip4«d br the Utter nop and held there br a eateh. The f*op F permits of there 
being a greater lei^h betveen the javs than i^ possible vhh the Flop E. «o that 
when the current u turned on a greater length o( material is heated. When the 
desired tempeiatare is readied the work 
» remored from the jaw? and aUowed 
to eocd eknrly. We may £av. with regard 
to the distance between the jaw? when 
weUii^. that the company enunciates 
as a rough-and-ready guide that gener- 
ally it may be taken that with round 
steel a proiection from the jaw? equal 
to double the diameter ol the material 
will be found to be most satisfactory. 

The butt and seam welding attach- 
mentii, of which mention has been made 
above, are exceedingly ingeniouf. As 
far as we are aware, the A 1 C<Hnpany 
is alone in the manufacture of these 
special attachments for apphcation to 
its standMd spot welding machines. It 
is, not unnaturally, rather proud of 
them, and it draws atteotioa to their 
atility in cases in which there is not 
more work of all kinds than can be 
effected with one machine. 

The butt welding attachment does 
not caU for detailed deeeription. For 
it, the ordinary stakes or arms of the 
machine are u^d. The attachment, as 
will be seen from Fig. 65, which shows 
it applied to a Xo. 4 spot welder, con- 
sists of two gripping jaws, one cf which 
fits on and can be clamped to the top 
stake, the other being fitted on the lower 
stake. When in position the clamping 
jaws come vertically one above the other 
and the pieces to be welded arc held in 
a vertical position. The machine is 
Fi«.65.-''Al»Bmt Welding Attachment. operated exactly as for spot welding. 

The jaw pieces, which are of massive 
construction, are not themselves water-cooled, as it is found that the water* 
circulating syHtem of the stakes is sufiEcient to keep the jaws sufficiently cool for 
intermittent working. The bulk of the metal in the two portions of the attacliment 
in brasK, but the actual jaws in which the work is gripped are made of a readily 
renewable piece of copper on that side through which the current flows and of steel 
on the side on which the clamping pressure is applied. The initial distance apart 
of the jaws can be varied from J in. to 2 in. by means of the milled nut on the 
vertical npindle seen at the right-hand side of the top jaw. 


The seam welding attachment must be deficribed at greater length. First of all, it 
may be Baid that it is made in various forms. There is, for Instance, one in which the 
welding operation is performed longitudinally straight in towards the body of the 
machine, and in it the top roller only is revolved. We show an example of .it in Fig. 66. 
In another form, which is for circular welding, the rollers are mounted at right angles 
to their carrying arms. In still another the lower arm is made right-angular for such 
work as the circular welding of the ends of tins. Finally, there is a form in which both 

Fio. 66.— "Al" Seam Welding Attachment. 

the bottom and top rollers are power driven. For thin work, such as -^g in. added 
thickness, it is not necessary for the lower wheel to be power driven, but with heavier 
work power driving is advisable in order to prevent " plucking " of the metal while it is 
at welding heat. 

We cannot spare the space to describe each particular form in detail, and we propose, 
therefore, to confine our remarks to the form illustrated in Pig. 66, which illustrates 
the seam weWer attachment known as No. S 4a as applied to a No. 4 spot welder. The 
latter machine is slightly different in construction from the spot welder described 
above, which in that known as No. 2. 

The general arrangement and working of the apphance may be readily understood 


from the illustration. The bottom stake of the spot welder is remDved and its place 
taken by an arm which carries the lower roller. The portion carrying the upper roller 
is slipped on to the upper stake, from which, of courne, the electrode has been removed. 
When in the correct position the appliance is securely clamped on the stake by means of 
the nut seen at the top. Power is appUed through a belt drive on to a shaft from which a 
short countershaft is also driven by belt. The countershaft, which has three speed 
cones, carries at its end a toothed bevel wheel, which meshes with a bevel wheel carried 
on a spindle at the other end of which ia a worm meshing with a worm wheel on the 
short spindle, which, in its turn, meshes with a worm wheel on the ahort'spindle carrying 
the upper wheel. 

The upper roller is attached to a large block of copper, and is water cooled, there 
being a thin sheet of copper between the revolving roller and the water reservoir. The 
lower roller is also water cooled. In the illustration given the water nozzles are shown 
clearly. In another form of the attachment the nozzles are at the rear of the lower 
roller carrier arm, and the water is sprayed through a small jet into a small tank or 
trough in which the lower half of the roller dips. 

In the form of attachment in which the lower roller is pow^r driven the drive is 
obtained by means of a chain from the upper shaft to a point at the ro6t of the lower 
arm, the chain then being taken along the inside of the lower arm to the roller. 

It will not be necessary to iUustrate or describe at any great length the firm's seam 
weldiftg machines proper. They are, in principle, substantially similar to the seam 
welding attachments, and they are operated in very much the same manner as are the 
spot welders, saving that pressure is kept on the pedal during the welding of the whole 
length of seam, the current of course being allowed to flow during that period also, and 
not being cut off as in spot welding. In the smaller machines only the top roller is 
revolved by power, but in the larger sizes both rollers are power driven. There is also a 
pedal-worked clutch, by means of which the rollers may bestarted or stopped. With 
the rollers not revolving a quick depression and release of the arm-cperating pedal will 
cause the formation of a " tack " weld, this being intended to do away with the neces- 
sity for a separate spot welder for " tacking.'* The welders are furnished with plug 
boxes having fast and slow plugs, as well as four additional plug holes, so that eight 
heating speeds are available. The top arm carrying the roller is, as in the spot welders, 
hinged at the welding level of the bottom roller, so that a straight pressure may be 
obtained on the weld, and no current flows through the hinge. The standard top arm 
is not water cooled, but the revolving roller is cooled as in the seam welding attachment. 
The standard bottom arm is cooled as in the alternative method referred to in connec- 
tion with the seam welding attachment, the roller actually revolving in a small water 
trough. The standard machine has 9 in. arms, but arms up to a maximum of 36 in. 
can be fitted. The maximum capacity for which the machines are built is 10 kilo- 
watts. The pressure applied to the work can be adjusted in the same manner as in 
the firm's spot welders. 

In addition to making welding machines, the firm supplies various accessories, 
including rotary converters, generators, switchboards, switches, etc. In particular, 
it makes a rotary converter for transforming direct current at 460 volts to alternating 
current at 230 volts. This machine, which is of 16 kilovolt-ampere capacity and runs 
At 1500 revolutions per minute, is designed to operate three of the firm's No. 3 welders. 



The question of the strength of electric welds is, naturaUy , of first-class importance, and 
it has in consequence received a considerable amount of study. The matter id of great 
complexity by reason of the varying conditions under which the process is carried out, 
as well as of the different systems of welding which are available. We shall not attempt, 
therefore, to discus^ it in all its bearings ; indeed, there are not sufficient data in exist- 
ence to enable such a course to be taken. Nevertheless, very valuable investigations 
have been carried out, both in this country and abroad, and we propose briefly to review 
some of them. We would preface our remarks, however, by repeating that, in our 
opinion, finality has by no means been reached, and that it would not surprise us if 
the figures, which will be found in what follows, were to be considerably modified when 
the whole subject has received the more extended research which it undoubtedly 

As might have been anticipated by aU who have had any experience of the methods 
of Lloyd's Register, which is always willing to give trial to novel methods, that institu- 
tion, as soon as the application of welding to ship construction was mooted, decided to 
carry out an exhaustive series of tests with a view to ascertaining whether a set of 
rules, under which electric welding might be employed in shipbuilding, could be formu- 
lated. The original tests occupied a period of some six months, and they resulted in 
the issuing in August, 1918, of '^ Tentative Regulations for the Application of Electric 
Arc Welding to Ship Construction." The actual regulations do not concern us at the 
moment, but the general nature of the experiments performed and the conclusions 
which they led to may be profitably given. This we are enabled to do by the courtesy 
of the Secretary of the Society, who communicated to us the decision of the Greneral 
Committee as soon as it was arrived at. We may say here that those who desire to 
follow the matter more closely than we are, by reason of space, enabled to do in the 
present instance, may do so by perusing a paper entitled " Experiments on the Applica- 
tion of Electric Welding to Large Structures," which was read before the Institution 
of Civil Engineers on March 11th last, by Mr. Westcott Stile Abell. 

It appears that the general scope of the experiments included : (a) Determina- 
tion of modulus of elasticity and approximate elastic Umit ; (6) determination of 
ultimate strength and ultimate elongation ; (c) application of alternating stresses 
with (1) rotating specimens, (2) stationary test pieces ; (d) minor tests, such as.(l) 
cold bending of welds, (2) impact tests of welded specimens ; (e) chemical and micro- 
scopic analysis. 

The tests were carried out on specimens which were as large as possible, particu- 
larly in connection with the static determinations of elasticity, ultimate strength 
and elongation, some of the test specimens being designed for a total load of just 
under 300 tons. The advantage of these large specimens was that the effect of work- 
manship was better averaged and the results were more comparable with the actual 
work likely to be met with in ship construction. For alternating stresses the specimens 
were relatively of small size. For the rotating test.pieces, circular rods^ mainly 

S.W. u 


machined from a welded plate, were used, the diameters selected being 1 in. and | in. 
These bars, about 3 ft. in length, were attached to a lathe headstock, and a pure 
bending movement in one plane was applied by means of two ball races to which 
known weights were attached. The material of the bar was thus exposed alternately 
to maximum tension and to equal maximum compression once in each revolution. 
The machine was run at about 1060 revolutions per minute. Bars of identical material 
were tried in pairs, one specimen welded and the other unwelded, and the number 
of revolutions before the specimens parted was observed for various ranges of stresses 
varying from ±15 tona to ± 6 tons. 

In the second series of alternating stress experiments flat plates of three thicknesses, 
viz. I in., I in. and ^ in., were used. These specimens were tried in groups of four, 
each group consisting of one plain, one butt welded, one lap welded and one lap riveted 
plate. The specimens, which were about 14 in. long by 5 in. broad, were clamped 
along the short edges, so that the distance between the fixed lines was 12 in. Each 
plate was also clamped, near the middle, to the end of a pillar, which by means of a 
crank arm was caused to oscillate and to bend the specimen equally up and down by 
adjustable amounts, the maximum total movement in any of the experiments tried 
being -j^ in. The machine was run at various revolutions — ^not exceeding 90 per 
minute— and the number of repetitions at which the specimen parted was obser^^ed. 

Minor tests of various kinds were undertaken, of which the principal had reference 
to the suitability of the welded material to withstand such bending and shock stresses 
as might occur in the shipbuilding yards. The experiments on bending consisted of 
doubling the welded plate over a circular block of diameter equal to three times the 
plate thickness, and comparing the results with those of the plate of the same material 
but unwelded. 

In the impact tests heavy weights were dropped from various heights on to the 
welded portion of a plate 5 ft. in length and 2 ft. 6 in. in breadth, the weld being across 
the plate parallel to tli^e shorter edge. The deflections were noted and the condition 
of the weld was examined after each blow. The chemical and micrographical examina- 
tion followed the ordinary practice. 

The results arrived at are summarised in the following : — 

(1) Modvlua of Elasticity and Approximate Elastic Limit. — (a) In a welded plate 
the extensions in the region of the weld are sensibly the same as for more distant por- 
tions of the unwelded plate. (6) With small welded specimens containing an appreci- 
able proportion of welded material in the cross-sectional area, the relation between 
extension and stress is practically the same, up to the elastic limit, as for similar 
unwelded material, (c) The elastic limit, or the limiting stress beyond which extension 
is not approximately directly proportional to stress, appears to be slightly higher in 
welded than in unwelded material, (d) The modulus of elasticity of a small test piece, 
entirely composed of material of the weld, was about 11,700 tons per square inch, as 
compared with about 13,500 tons for mild steel and about 12,500 tons for wrought 

(2) Ultimate Strength and Ultimate Elongation, — (a) The ultimate strength of welded 
material with small sp^imens was over 100 per cent, of the strength of the unwelded 
steel plate for thicknesses of J in., and averaged 90 per cent, for plates of f in. and 
1 in. in thickness. (6) Up to the point of fracture the extensions of the welded speci- 
mens are not sensibly different from those of similar unwelded material, (c) At stresses 
greater than the elastic limit, the welded material is less ductile than mild steel, 
and tlfe ultimate elongation of a welded specimen when measured on a length of 


8 in. only averages about 10 per cent., as compared with 25 to 30 per cent, for 
mild steel. 

(3) Alternating Stresses. — (a) Rotating specimens (round bar) : (1) Unwelded 
turned bars will withstand a very large number of repetitions of stress, exceeding, say, 
5 millions, when the range of stress is not greater than from 10^ tons per square inch 
tension to 10^ tons per square inch compression ; (2) welded bars similarly tested will 
fail at about the same number of reversals when the range of stress exceeds ± ^ tons 
per square inch, (b) Stationary test pieces (flat plate) : (1) Butt welded specimens will 
withstand about 70 per cent, of the number of repetitions which can be borne by an 
unwelded plate ; (2) lap-welded plates can endure over 60 per cent, of the number 
of alternations necessary to fracture a lap-riveted specimen. 

(4) Minor Tests, — (a) Welded specimens are not capable of being bent, without 
fracture, over the prescribed radius to more than about 80 deg. with ^ in. plate, reduc- 
ing to some 20 deg. where the thickness is 1 in. Unwelded material under the same 
conditions can be bent through 180 deg. (6) Welded plates can withstand impact with 
a considerable degree of success ; a | in. plate, 5 ft. in length by 2 ft. 6 in. in width, 
sustained two successive blows of 4 cwt. dropped through 12 ft., giving a deflection of 
12 in. on a length of about 4 ft. 6 in. without any signs of fracture in the weld. 

(6) Chemical and Microscopic Analysis. — (a) Chemical analysis : (1) The electrode 
was practically identical with mild steel, but there was a greater percentage of silicon ; 

(2) the material of the weld after deposition was ascertained to be practically pure 
iron, the various other contents being carbon .03, silicon .02, phosphorus .02 and 
manganese .04 per cent, respectively. (6) Microscopic examination : (1) The material 
of the weld is practically pure imn ; (2) the local effect of heat does not appear largely 
to affect the surrounding material, the structure not being much disturbed at about 
-^ in. from the edge of the weld (the amount of disturbance is still leds in thin plates) ; 

(3) the weld bears little evidence, if any, of the occurrence of oxidation ; (4) with welds 
made as for these experiments, i.e. with flat horizontal welding, a sound junction is 
obtained between the plate and the welding material. 

(6) Strength of Welds {Large Specimens). — (a) Butt welds have a tensile strength 
varying from 90 to 95 per cent, of the tensile strength of the unwelded plate, (b) Lap 
welds : (1) With full fillets on both edges the ultimate strength in tension varies from 
70 to 80 per cent, of that of the unwelded material ; (2) with a full fillet on one edge 
and a single run of weld on the other edge the results are very little inferior to those 
where a full fillet is provided for both edges, (c) Riveted lap joints : For plates of 
about i in. in thickness, the specimens averaged about 66 to 70 per cent, of the strength 
of the unperforated plate. 

There are two points in connection with the foregoing which call for mention. 
The first is that the experiments only referred to work effected by arc welding, and 
the second that only one particular form of coated electrode was employed. In this 
connection it may be mentioned that the '^ Tentative Regulations " provide that 
(a) the process of manufacture of the electrodes must be such as to ensure reliability 
and uniformity of the finished article ; (6) specimens of the finished electrodes, with 
specifications of their nature, must be supplied to the Committee ; (c) that the Com- 
mittee's officers shall have access to the works where the electrodes are made, so as 
to ensure that the electrodes produced are identical with approved specimens ; and 
(d) that no alteration in the method of manufacture may be adopted without the con- 
sent of the Committee. It is quite obvious, therefore, that the Committee attaches 
great importance to the question of electrodes. 

Early in last year Captain (now Major) James Caldwell, R.E., Deputy Assistant 


Director of Materials and Priority at the British Admiralty, was, at the request of the 
United States Shipping Board, lent to the Emergency Fleet Corporation. At the 
conclusion of three months* service Captain Caldwell issued a voluminous report on 
** Electric Welding and its Applications in United States of America to Ship Construc- 
tion/' Included in the report is a lengthy table setting out the results of an extensive 
series of tests carried out under the direction of Mr. W. T. Bonner, of the Chester 
Shipbuilding Company. In order to be fully appreciated, the whole of the information 
given in the table should be taken into consideration, but, as space will not permit 
of us to reproduce the table in extenso — ^it would take a page and a half of the Engineer, 
and even then be reproduced to a small scale — ^we must content ourselves by giving 
the following summary. The starting-point is an ordinaiy single-riveted lap joint 
effected in accordance with Lloyd's regulations, one seam being caulked. The thickness 
of the plates employed varied from .234 to .30 square inch, the width from 9 in. to 
9|ij^ in., and the area involved froxa 2. 106 square inches to 2.719 square inches. The 
mean of a series of tests made on joints of this type and size showed that the riveted 
plate had a tensile strength of about 52 J per cent, of the strength of an unperf orated 
plate of like size and form. For simplicity's sake we will call this joint A. A double- 
riveted joint, singly caulked, of approximately the same dimensions was found to 
give a mean of 76^^ per cent, of the strength of an unperforated plate. A joint similar 
to A, but with one seam welded instead of caulked, gave as a mean a strength of 71^ 
per cent of the unperforated plate. With a similar joint, but with a full fillet weld, the 
percentage rose to 75^-. A lap joint of veiy similar dimensions to A, but with a fillet 
weld at both plate edges and no rivets, gave as a mean of several tests a strength of 
94 per cent, of unperforated plate of equal dimensions. A butt-Welded joint furnished 
with a strap fiUet-welded at both edges, but no rivets and having dimensions com- 
parable with those of joint A, gave as the result of several tests a mean strength of 
95^ per cent, of that of unperforated plate. The foregoing tests were apparently 
made with the joints just as they were left by the welder, for later in the table are 
given side by side examples of butt joints before and after being machined. The joint 
before machining gave in the two cases recorded a strength equal to 100 per cent, of 
that of unperforated plate, while the two cases of machined joints showed strengths 
of 70^^ and 77^ per cent, respectively. 

In the case of all the foregoing arc welding was employed, but there is no indica- 
tion as to what type of electrode was used. There is only one set of tests in which 
the specimens were spot welded. The joint was of the lap type, had a length of 10 in. 
and an area of 3.08 in., the thickness of the plate being 0.308 in. The strength of 
the joint was found to be only 6 per cent, or under of the unperforated plate, and the 
remark is made the efficiency was " too low for serious consideration." This is inter- 
esting in the Ught of the results which we shall now refer to. 

In the April, 1919, number of the Proceedings of the American Institute of Electrical 
Engineers there is printed a paper on " Welding Mild Steel," which was read by Mr. 
H. M. Hobart at a joint session of the Institute with the^ American Institute of Mining 
Engineers held at New York on February 20th, 1919. The paper deals principally 
with the investigations undertaken by the Welding Research Sub-committee of the 
Welding Committee of the Emergency Fleet Corporation, and it forms a valuable 
contribution to the literature on the subject. With the bulk of it we are not at the 
moment concerned, but there is one portion of it which deals with the strengths of 
various types of joints that bears directly on .the matter which we have under con- 
sideration. Mr. Hobart points out that a great deal of progress is being made in 
America in the use of spot welding as a means of joining thick plates. It is believed. 



he remarks, that spot welding has a great future as applied to shipbuilding, and 
several large welders have been built for shipyards.* "In some of its applications/' 
he continues, '^ spot welding affords a method of preliminarily jointing the hull 
plates, after which the additional strength is provided by arc welding. The Welding 
Research Sub-committee has already made some progress in comparing combined 
spot and arc welds and combined rivet and arc welds with riveted, spot-welded,, 
and arc-welded joints. It is not a question, in such an investigation, of spot versus 
arc welding, but of cpot and arc welding." 

In the tests alluded to the specimens were made up of the following combination : — 

(1) Spot and welded fillet, as in ^ 

A in the^ accompanying diagram, /r^^^~y^''yji j^r^^^T P^ 

Fig. 67. // f "\ f ,f-^ / y 

(2) Fillet welded, made by 

welding fillets about 2 in. in length A- Fillet and Spot Welded 

at the ends of the plates, as in B. 

(3) Riveted and fillet welded, 
as in C. 

(4) Spot welded, made by weld- 
ing two spots approximately 1 in. 
in diameter on the plates, as in D. 

(6) Riveted joint made by rivet- 
ing a J in. by 4 in. by 12 in. plate 
with two plates \ in. by 4 in. by 
16 in., using two | in. rivets and 
a 4 in. lap. 

The ultimate loads borne by these 
joints were as follows : — 

(1) Spot and fillet welded, 60,360 lb. 

(2) Fillet welded, 37,000 lb. 

(3) Riveted and fillet welded, 

35,000 lb. 

(4) Spot welded, 28,000 lb. 

(5) Riveted joint, 13,000 lb. 


B -Fillet Welded 

C —Riveted and Fillet Welded 

I) -Spot Welded 

£ — Riveted Joint 
Fig. 67. — Welded and Riveted Joints. 

It is not quite clear why No. 3 should be less than No. 2, but possibly were a larger 
number of test pieces to be made than was actually the case the figures might show 
some modification. A noticeable feature is the strength of the spot-welded as compared 
with the riveted joint. It would seem likely that in the tests made under the super- 
vision of Mr. Bonner, and quoted above, the strength of the spot welding machine 
ufied was not great enough to deal adequately with plates of the thickness which he 
was employing. 

The foregoing embodies some of the latest published data regarding the strength 
of various types ol electric welds. In it, however, it has been possible to do no more 
than touch upon the fringe of what is really a very wide subject. It has to be realised 
that in electric welding, and especially in arc welding, the personal element counts 
for a great deal. It is, unfortunately, by no means easy to ascertain by inspection 
after it has been made whether a weld is good or not. An efficient and experienced 

* There is little doubt that Mr. Hobart was referring to the machines described in Chapter XI. 


arc welder wiU be able to tell as he is making the joint whether the joint is good or 
bad. The inexperienced welder may not be able to do so, and he may yet produce 
work which to outward seemiiig is good. As we pointed out in an earlier chapter, 
however, there is no guarantee that a welded joint made by a smith in the ordinary 
way is sound throughout. A good smith will make it so. While an inefficient smith 
may leave impurities between the surfaces he is trying to weld. In fact, it may be said 
that after taking everything into consideration the evidence available seems to in- 
dicate that in the various forms of electric welding the predent-day engineer has at 
his command methods of joining metals together which, if properly applied, are in 
most cases at least as efficient as, and in some instances more efficient, than any which 
have gone before. In saying this we do not wish to be taken as minimising the value 
of acetylene welding. On the contrary, we are satisfied that, in some directions, the 
acetylene torch is the most efficient tool which can be employed. We are convinced, 
however, that a good deal of weeding out is required and that a not inconsiderable 
amount of improvement of existing methods will have to be effected before the full 
benefits derivable from electric welding can be experienced. Such weeding out is 
bound to be done, however, and such improvements are certain to be brought about, 
so that the engineer of the future will most certainly possess more perfect methods 
than those which exist at the present day, excellent as, in most ways, they are. 



In that chapter of the series on " Electric Welding and Welding Appliances " in which 
we dealt with the machines and processes of Pontelec Welding Patents, Limited,* 
we made mention of having seen the drawingls of a large spot-welding machine which 
was, at the time the chapter was written, under construction. The maohine has now 
been completed, and it is believed by its makers to he the largest spot welder o£ its 
type which has up to now been constructed for commercial purposes in this country. 

Fia. 68.— Pontelec Spot Welder, with Stakes ar»nged horizontalty. 

The particular work which it was designed to perform is the spot welding of tubes in 
6 ft. lengths mode of { in. steel plates, and having an internal diameter as small as 8 in. 
Naturally, however, larger diameters can be dealt with. In the two views of the machine 
which are given in Figs. 68 and 69 it will be observed that the design is such that the 

• 3eeChftpt«rV. 


etakefl or anna can either be used in a vertical or a horizontal position. The body of the 
machine ift hinged to the bed by bolts, one of which, with its nnt, oan be seen at the 
left top corner of the nde frame in both Ulustrations. When the Bt^es are horiz<»ital 
the body is held in poeition by means of two eye-holed pinB, which pass throngh boles 
in the side frames and enter similw holes formed, one on cme side and the other on the 
other of the framework of the body of the machine. When it is desired to have the arms 
vertical, the weight of the body of the machine is taken by means of a crane, the hook of 
which is passed through the eye-bolt seen at the top. The supporting pins can then be 
withdrawn, and the body of the machine is allowed to hinge down ao that it enters the 
space between the side frames of the 
base. When the stakes are vertical, 
holes in the body of the machine-~one 
of which is seen at the top, just to the 
left of and below the eye-bolt in Fig. 68 
— register with holes in the side frames 
— one of which b seen at the right top 
comer of the side frame in Fig. 68 — 
and the pins, which were removed just 
before lowering commenoed, can be 
inserted. When in either the horizontal 
or vertical positions the weight of the 
body is taken on four pins, one at each 
comer. The pins are 2 in. in diameter, 
and hence of ample strength to support 
the heavy weight of the body. The 
total weight of the machine is, we may 
say here, in excess of 4^ tons, some 
H tons of which is represented by copper. 
The construction of the machine will 
be readily seen from the illustrations. 
The bed is made up of two side plates 
and two end plates, all of cast iron, and 
all bolted seeurely together. The body 
part, which supports the stakes or 
electrode arms and contains the trans- 
former, is composed of cast iron end 
Fio. 69, pieces separated by and bolted to cast 

The Welder, with Stakes arranged verticaUj. iron pillars, which are star-shaped in 

cross section. The transformer core is 
built between and is supported by these pillars. The end pieces have each cast in the 
centres of their ends semicircular chaimels, which are more or less in the form of 
plummer blocks, and which are machined to a radius of 3| in. Bo as to receive the 
stakes. The latter, which are 6| in. in diameter, are secured in position by means of 
caps bolted on in the same manner as are the caps of bearings. The caps at the front 
side of the machine — that is, the side from which the stakes project — are of copper, and 
to them are connected the terminals of the secondary winding of the transformer. The 
ca{)s at the back side are of cast iron, since they are not required to carry any current. 
The primary of the transformer is fomished with ten tappings, so that a wide range 
of welding speeds is obtainable. 

As it was necessary for the machine to have such a wide reach as 5 ft., special 


consideration had to be given to the construction of the stakes. It was realised from 
the outset that pure copper alone would not give the requieite stiffness, but would 
have excessive " spring." On consideration, it was decided to use cores of manganese 
steel with electrolytic copper, containing a small amount of silicon, oast on them. For 
the cores Hadfield's Hecla manganese steel bull-headed railway rails, weighing 95 lb. 
per yard and rolled in accordance with the British Standard Specification, were em- 
ployed. As showing the degree of stiffening imparted to the arms by the rail rein- 
forcement we may say that we understand that with a load of from 10 cwt. to 1 5 cwt. at 
5 ft. extension the " spring " does not exceed J in. The arrangement of both top and 
bottom arms is shown in Fig. 70. Both fums are of the same diameter, and both are 
machined so that they can be slid backwards and fOTwards in their bearings when the 
caps of the latter are slacked off. The top arm has an extreme length of 7 ft. 3 in., and 
is parallel from end to end . The rail reinforcement is taken to within 3 in. of the extreme 
forward end, and it projects 1 in. at the rear end. The lower arm has an overall length 
of 7 ft. 9J in. Its sides are parallel for a length 5 ft. 1 in. from the rear end, and in the 
remaining 2 ft. 8^ in. the sides of forward portion are machined off taperwise until, as 

Fis. 70. — Upper and Lower Stakea or Arma. 

may be ^een in the drawing, the distance between them is only 4 in. Two inches from 
the extreme front end a vertical hole 2 in. in diameter is drilled for a depth of 4} in. 
for the reception of renewable eleotrodes, the projection of which above the top of the 
arm can be adjusted by means of a set screw 1 in. in diameter, for which a tapped hole 
is formed concentrically with and at the bottom of the hole containing the electrode. 
A ^ in. slot is cut vertically in the end of the arm for « depth of 6^ in., and a horizontal 
hole is drilled in the arm at a distance of 4 in. from its end, so that by means of a bolt 
and nut the electrode may be firmly gripped. The rail reinforcement is only taken to a 
point 9^ in. of the front end, the remainder of the arm beii^ copper. There is a pro- 
jection of 1 in. of the rail at the rear end, oe in the upper arm. The lower arm has two 
grooves 1 in. wide and I in. deep, with their bottoms curved to ^ in. radius, formed one 
on each side of the horizontal diameter. These grooves are intended for the accommo- 
dation of pipes for carrying water for cooling purposes should they be found necessary. 
We gather, however, that, so far, artificial cooling has not had to be resorted to, the 
large mass of copper in the fums having proved ample for the conduction away of the 
heat during working. It should be mentioned, though, that up to the present the 
machine has not been worked to its full capacity. The maximum energy at the dis- 
posal of the Ponteleo Company for testing purposes is otdy 50 kilovolt-amp^res. With 


that amount of energy we understand that the maohine has satisfactorily welded jt m- 
plates — or an added thicknesa of J in. — with the arms fully extended. It is thought 
probable, however, that when sufficient power is available the machine will readily 
weld two ^ in. plates — or an added thickness of 1 in. The current regulation is such 
that, at the other end of the scale, sheets as thin as No. 22 gauge can be dealt with. 

The makers point out that it is not, of course, necessary that the electrode-carrying 
arms should be circular in section. In the present case the form was practically settled 
by the fact that the tubes, which the machine was specially designed to weld, were 
only 8 in. in diameter, and because it was desired to have the gap adjustable in height. 

"The ENCi.tEll" SwA.-. St T-l EMi^tin" »— « St 

Flo, 71. — Arrangement of Electrode Slide. Fio. li. — Arrangement of Transformer. 

It will be understood, of course, that by slacking away the caps holding the arms in 
position the length of projection of the latter can be altered at will. We gather that 
the minimum projection gives a free-reach of only about 9 in. 

The upper electrode arm remains, as may be gathered from the illustrations, rigid, 
and is nob moved up and down, as in some forms of spot welders. It is the electrode 
which is reciprocated, and the means by which the reciprocation is effected is clearly 
shown in Fig. 71. It will be observed that the electrode A, which is 2 in. in diameter, 
is inserted in a hole drilled in the extremity of a soUd copper sliding piece B. The 
latter, which has, in cross section, the form of a truncated cone, fits into and can be 
moved up and down or backwards and forwards, as the case may be, in a slide of 
exactly similar shape as itself, which is formed by a body C and a cap D. the latter 
being securely attached to the former by means of screwed bolts. The sliding piece. 


body, and cap are of copper, the studs of brass, and the nuts only of iron. The 
extremity of the upper electrode arm E fits into a boss F fonned on the body C, and, 
since there is a slot G cut in the periphery of the latter, the electrode arm can be 
gripped tightly by means of the bolt and nut H. 

Housed in a channel formed in the sliding piece C is a spring J, and arranged in a 
channel 2| in. wide and | in. deep cut in the sliding piece is a toothed rack K, which 
is furnished at its upper end with a right-angled extension L. The latter is bored and 
tapped to receive the screw M, at the end of which is an enlarged portion N, which 
impinges on the top of the spring J. By means of this arrangement any desired 
initial compression may be put upon the spring. Meshing with the toothed rack is a 
toothed pinion O, which is keyed to the shaft P. The shaft can be rotated by means of 
the lever seen in Figs. 68 and 69, the baleinced weight on which is so arranged that the 
tendency is always to draw the upper electrode away from the lower. It will be under- 
stood that the effect of moving the lever is to push the electrode down on to the work, 
further depression giving greater conapression to the Spring, and hence greater pre^ure 
on the work. The makers point out that, of course, sway other mechanical means of 
exerting the pressure necessary for welding, as weU as compressed air or hydraulic 
pressure, can be applied to take the place of the mechanism illustrated. 

The winding of the transformer is for 50-period alternating current at 160 volts, 
though, naturally, any other winding to suit different conditicms can be applied. The 
primary circuit is closed and opened by means of a foot-operated switch. In Fig. 72 
we show the outline of the arrangement of the transformer for a machine of similar that described in the foregoing, but with a permanent base for horizontal work 
only. The electrical connections are, however, identical in both cases. 



The Plastic-arc welding system is the name given to the process of electric arc welding 
which is controlled by the Wilson Welder and Metals Company, of 2 Rector Street, 
New York City. It was by means of it that the German steamships which were in- 
terned in harbours in America, and. which had been wilfully and seriously damaged 
by their crews, were repaired. Before describing the system itself, it will perhaps 
be of interest to our readers if we briefly refer to the manner in which the vessels were 
damaged and to the steps taken to repair the breakages. 

Oh the declaration of war the United States authorities took over all the enemy 
ships which had been interned in various ports in the country. There were 103 vessels 
in all. In the port of New York alone there were interned 31 steamships, of which 
27 were Grerman and 4 were Austrian. Of the 27 German vessels 2 were sailing ships 
and 4 were small steamers. The latter had evidently been considered by the Germans 
to be of comparatively small importance, for the destruction wrought in them was but 
slight ; but the remaining 21 vessels, which ranged in size from the 56,000 tons or so 
of the Vaterland to the 3900 tons of the Nassovia, had been so seriously damaged that, 
when taken over by America, they were entirely valueless for transport purposes. The 
Germans, with the thoroughness which characterises them even when engaged in work 
of destruction, had so mishandled the machinery that in no case was there an engine 
which could turn round. A survey showed that of the 20 vessels in New York Harbour 
which were fitted with reciprocating engines — ^the Vaterland ik turbine-driven — there 
were 118 fractures of an important character, which would, had ordinary repair 
procedure been followed, have necessitated the replacement of seventy steam cylinders. 
As a matter of fact, it was at first proposed to proceed with such replacement, but 
when the question had been further investigated it was realised that the length of 
time required to carry out the repairs in that manner would be prohibitive if the ships 
were to be used for the transport of troops across the Atlantic so ad to be of any real 
service in the war. The principal damage done, according to a report made by the 
Secretary of the Navy, was the breaking of cast iron parts of the main engines — 
chiefly the cylinders — ^though in one case piston-rods, connecting-rods, and boiler 
stays were sawn half-way through, and in others the boilers were ruined by dry firing. 
There was, in addition, much vandalism of a minor character, but the task of remedy- 
ing it was insignificant in comparison with the gigantic business of repairing the 
cylinders, some of which were more than 9 ft. in diameter. The methods of destruction 
adopted were various. Apparently dome sort of a battering-ram bad been exten- 
sively employed, and if that apparatus were not by itself capable of doing what was 
required of it numerous holes were drilled, in some cases a complete line of holes 
being made round the part at which the battering-ram was being appUed. The different 
kinds of damage were, however, too numerous to be described in detail in the present 
instance. We can only refer to the engravings Figs. 73 to 76 inclusive, which are 
reproductions of certain photographs of different pieces of machinery in various vessels 
taken before the work of repair was commenced. 


Via. 73. — S.S. Gtorge I^athington — Fractured Circulating-Pump Casing. 

Fio. 74.-S.S. Frinz /oacftim— Broken Cylinder-Head and broken Low-Preadure Cylinder. 


Fio. 76.— S.S. Kaiter WiUttlm /7,— Fractured Low-Pressure Cylinder Liner. 

Fro. 76.— S.S. FriedHch der (Trom— Broken Steam InUke Second Intennediate Cylinder 


The credit of having suggested that the repairs might be effected by means of arc 
welding is attributed to Capt. E. P. Jessop, of the United States Navy, the Engineer 
Officer of the New York Navy Yard, to which place a few of the vessels were trans- 
ferred for repair in the first instance. His recommendation was readily endorsed by 
Bear- Admiral Burd, the industrial manager of the yard. The matter was referred to 
the Bureau of Steam Engineering, and Capt. O. W. Koester, assistant to the Bureau, 
was directed to make a thorough examination of all the conditions on the ex-German 
ships. As a result of his investigation, orders were issued to make all the necessary 
repairs, where possible, by electric welding, and to resort to mechanical patching only 
where welding was impracticable. 

It is explained by the Secretary of the Navy, in his report for the year ending 
December 1st, 1918, that the '^ decision, so far-reaching in its application and so fraught 
with danger to the professional reputation of the officers concerned," was come to in the 
face of opposition from engine builders and marine insurance companies, but that it 
was made " with such confidence in the ultimate result as left no room for doubt of 
its success." Yet, though, of course, electric welding had been in successful operation 
for some considerable time in connection with ship repairs, it had never before been 
employed on such an extensive scale as was then proposed. It was realised, however, 
that if it could be successfully applied, it would not be necessary to remove the cylinders 
from the vessels, and it was calculated that the ships would be ready for service 
months earlier than would be the case if the cylinders were taken out and replaced, 
and, further, that there would be a great saving in money. 

The decision was fully justified by the results attained. " So well and so success- 
fully were the repairs accomplished that," states the Secretary of the Navy, " there 
was not a single instance of a defective weld, nor has one developed during the months 
of arduous service on which these ships have been engaged." And he continues, '' as 
a sidelight on this work it has been made the subject of careful estimate and deter- 
mination that the repair of these ships resulted in the saving qi twelve months in time, 
enabled us to transport at least 500,000 troops to France, and effected an economy 
which is conservatively estimated at upwards of 20,000,000 dols." 

We understand that the cylinders of fifteen of the vessels were repaired by electric 
welding, while those of six were repaired by fitting mechanical patches. . In other words, 
eighty -two of the major injuries were repaired by welding and thirty-six by mechanical 
methods. We are informed that the whole of the welding work in connection with, 
the vessels shown in the accompanying table (see page 112) was effected on the Wilson 
system, which we shall now proceed to describe. 

The first sixteen vessels in the above table, representing over 288,000 gross tonnage, 
were made available for the transport of American troops and supplies overseas ^within 
the space of five and a half months. The first two ships to be tackled — ^the Friedrich 
der Grosse, now the Huron, and the Prinzess Irene, now the Pocahontas — were ready 
for sea within two months from the time that the repairs were undertaken, in spite 
of the fact that their engines were among the worst damaged of aU — ^so badly, indeed, 
that with ordinary -methods nine cylinders would have had to be replaced. 

The sole agents in this country for Plastic-arc welding plant and apparatus are G. D. 
Peters and Co., Limited, of 15 Dean's Yard, S.W. 1, and of Windsor Works, Slough. 
At the latter we recently had an opportunity of witnessing the system in operation 
within a day or two of its having been got to work. The current was supplied by an 
ordinary motor generator set, the generator being compound wound with interpoles 
and designed to generate direct current at 37 volts. There is nothing special in either 
the motor or the generator, nor are they electrically connected. The gist of the system 



lies in the control panel, which is illustrated in Fig. 77, and the connections of which 
are shown in Fig. 78. The panel is equipped for both welding and cutting, but it is in 
the connections for welding that its chief interest lies, since, when used for cutting, 
only the reactance coil, the resistance grid shown at the left-hand bottom corner of 
the illustration, and the ammeter are in circuit. When, however, the connections 
are set for welding — ^that is, when the double-pole knife edge switch is, as shown 
in the left top comer of the engraving. Fig. 77, thrown over to the right-hand contacts 
— other pieces of apparatus, which differ from those of any other system with which 
we are acquainted, are brought into play. On tracing the connections shown in Fig. 
78, it will be observed that the positive terminal of the generator is connected directly 
— and without passing through any apparatus saving the knife switch and plug switch 
E — ^to the work. The negative pole, on the other hand, runs from the knife switch 


German name. 

Grosser Kurf urst 

Kaiser Wilhelm II 

Amerika . . 



(Jeorge Washington 
Friedrich der Grosse 


Koenig Wilhelm II 
Martha Washington 


Kronprinzessin Cecelie . . 

Prinzess Irene 


President Grant 

President Lincoln 




U.S. name. 






George Washington . . 




Martha Washington . . 


Mt. Vernon 


President Grant 
President Lincoln 


Susquehanna . . . . 
























• Transport. 

t Repair ship. 

X Shipping Bd. 

first of all to the solenoid A, Fig. 77, which may be termed the brains of the apparatus. 
The plunger of the solenoid is furnished at its upper extremity with a graphite piston 
which works in a dashpot B, and so controls the motions of the plunger that hunting 
is prevented. From the solenoid connection is made to the right-hand side of a re- 
sistance C and from the left-hand side of the latter through the ammeter D to the 
opposite terminal of the plug switch E to that occupied by the positive lead. The 
resistance C consists of a series of carbon plates which are plain on one side and pro- 
vided with numerous regularly placed projections, carefully ground to the same height, 
on the other. The amount of resistance offered to the passage of current through this 
pack of carbon plates can be varied by the amount of jressure put upon them by 
means of the lever F^ which is pivoted at F and is rocked up and down on that point 
by the movements of the plunger in the solenoid. If the current in the welding circuit 
tend to increase beyond a predetermined value, then the plunger, and with it the 
lever F^, are raised, with the result that the pressure on the carbon plates is relieved, 

iSls 1, 



and in that way additional resistance inserted into the circuit. Then, if the other con- 
dition^ remain the same, the current flowing in the circuit will fall, since the supply 
voltage is constant. Conversely, if the current tends to fall below the required value, 
the plunger and with it the lever F^ will also tend to fall, with the result that the 
resistance of C is reduced by the compression of the carbon plates, and the current 
flowing in the circuit will rise. 

The quantity of current to be allowed to flow can be varied at will by what is 
equivalent to running a weight backwards and forwards along the lever F, so that 
the eflEort to raise it required of the solenoid is decreased or increased. As a matter 
of fact, instead of a weight, the required effects are actually produced by decreasing 
or increasing the tension on two coiled phosphor-bronze springs, the front of which, 
G, is clearly shown in the engraving, that at the rear being hidden by its companion 
in front. These two springs are connected to two small roller trolleys, one of which 
can travel on the top of the lever F^, while the other impinges on the lower edge of 
the fixed inclined guide H. It will be readily understood that since the guide H is 
inclined downwards from the resistance C, the further the springs G are away from 
C the greater will be the effort required of the solenoid to lift the lever F^ and hence 
to reduce the pressure on the carbon plates. It follow^, therefore, that the greater 
the distance of the springs G from C, the larger the current will be before the solenoid 
will begin to take control. Conversely, the nearer the springs G are to C, the less the 
effort required of the solenoid and the smaller the current flowing in the circuit before 
the solenoid begins to act. Hence, by varying the position of the springs G, the exact 
working current can be arranged for before operations commence. For the initial 
adjustment of the carbon plates there is a screwed bolt at the left-hand side. This 
bolt, however, will only require to be used for initial adjustment, and not in ordinary 

In some forms of the Wilson equipment this adjustment of the desired working 
pressure on the carbon plates is effected by a hand wheel and screw, but in the equip- 
ment which we saw, and which is shown in Fig. 77, the screw J, which passes through 
a threaded member forming an integral part of the spring-carrying trolley that travels 
on the top of the lever F^, can be revolved in one direction or the other by the revolu- 
tion, in one direction or the other, of the small vertical spindle motor K. The advantage 
of the latter arrangement is that if the operator finds on commencing to work that the 
current flowing in his arc is either in excess of or smaller than the quantity required 
to do the work to the best advantage — ^and he becomes aware of the ruling conditions 
by the aspect of the work in progress — he can either decrease or increase it at will 
by moving the motor control switch — see Fig. 78 — either to one side or the other. 
The switch may be arranged so as to be portable and to be carried from one place to 
another, so that the required alteration in the working conditions can be made without 
the operator having to make a special journey to the control board for the purpose. 
It may be added that there is a post G^ rising vertically from the trolley running on 
the Isver F, which can engage with and can open two carbon block switches, which are 
normally kept closed by spring action. One of these switches is arranged to be opened 
at one end of the travel of the trolley, and the other at the other, so that if the trolley 
should tend to overrun its travel in one direction or the other the circuit of the motor 
K is opened. 

In actual operation the movements of the plunger in the solenoid, and hence of 
the lever F^, are not large, and the pointer of the ammeter keeps remarkably constant. 
Still, there is a continuous and quite clearly visible compression and decompression 
of the carbon blocks with the variation of the arc at the weld. All that happens if 


the electrode is kept on the work is that the resistance C gets hot, and would un- 
doubtedly burn if the short circuit were continued long enough ; but we understand 
that no damage would be done to the generator, since the resistance offered by the 
carbon plates would be enough to preclude the passage of a current sufficiently heavy 
to injure the machine. 

We mentioned earlier in this chapter that the initial voltage was 37. In some 
equipments at work in the United States the initial voltage is some two volts lower 
than that. In the equipment that we inspected the voltage drop across the resistance 
when welding was being performed was about 20. Hence for maintaining the arc and 
for overcoming the resistance in the leads and solenoid there was only available 17 
volts, and that is all that is required. The electrode being employed when we saw the 
plant in operation was plain iron wire about ^^^ in. in diameter. It was not coated, 
nor was flux of 'any kind employed. The work being done was, to all appearances, 
excellent, and from tests, the data of which were shown to us, there is no doubt that 
the system is capable of doing work of the very highest quality, the consequence, 
it is explained, of having the current " just right " for each particular operation, and 
of the ability to adjust it to the desired quantity without difficulty. The critical heat 
at which the particular metal being welded should be fused is thus kept constant 
at the arc, and there are neither voids nor burning. With an electrode of the size 
mentioned — f^ in. — we are informed that the amount of metal deposited by One operator 
per hour with a heat value of from 90 to 140 amperes at from 35 to 40 volts ranges 
from I lb. to as much as 2J lb. to 3 lb., the average being about 1 J lb. per hour, the 
actual quantity depending on the character of the work being carried out. Further- 
more, we gather that the number of lineal feet that may be welded by one welder 
ranges from 1^ ft. on | in. plate, with a liberal overlap of metal, to as much as 10 ft. 
per hour on J in. stock, with no extra metal applied, a safe average being probably 
about 4 ft. per hour. The results of tests which have been shown to us demonstrate 
that the average tensile strength per square inch at the weld arrived at with eleven 
specimens having an area of 0.628 square inch, the weld being planed flush with the 
plate, was 54,700 lb., the elongation in 2 in. averaging 7.67 per cent. Nine specimens 
having an area of 0.572 square inch gave 56,700 lb. as tensile strength and 10.0 per 
cent, elongation in 2 in., while eleven further specimens having an area of 0.595 square 
inch gave 58,900 lb. and 7.53 per cent, respectively, the figures being avei*age in all 
cases. These results were obtained by taking a rectangular boiler plate 18 in. by 20 in. 
of known tensile strength, cutting through the centre the long way and welding with, 
a quality of wire suitable for the purpose. The plates were then cut into test pieces 
1| in. wide and tested in the regular way after the welds had been planed flush on 
both sides. It is claimed that welds made by the Plastic-arc system will stand much 
more in the way of twisting and bending than will welds made by any other system. 

The Wilson system was in the first instance, and long before it was employed for 
ship repairs, developed for railway work. A very large volume of railway repair work 
has been done by arc welding in the United. States for a good many years past, and has 
been found to be convenient, effective and cheap. As an example of the class of job 
which is frequently met with, we may mention cracked locomotive frames. Wrought 
steel is largely used for locomotive frames in the United States, and it has been found 
that when cracks develop in them the damage can readily be repaired by electric 
welding.- The engraving, Fig. 79, shows a broken frame of engine No. 1653 of the 
Grand Trunk Railway in process of being repaired by the Plastic-arc system at Strat- 
ford, Ontario. The method employed is to cut away the metal, as shown, on each 
side of the frame in the neighbourhood of the crack by means of pneumatic hammers. 


The space left by the material cut away is then filled by a piece B of metal 
identical with the metal of the frame and square in ctobb section, and with two 
. opposite comers matching with the V-shaped ends of the frame formed as a result 
of the metal being cat away. Weldii^ is then commenced at the bottoms of the V- ' 
shaped cavities formed on each side of the frame, as shown in Fig. 79, and the process 
is continued until the cavities are filled up and their surfaces brought fiush with or 
perhaps a little proud of the original surface of the frame, as shown in Fig. 80. Repairs 
such as that just described are frequently carried out, and are found to be quite effec- 

Figs. 79, 80. 
Cracked LocomotiTe Frame— Early Stage and Final Appearance of Repair by Arc Welding 



As has been pointed out in an earlier chapter, direct current is almost exclusively 
used for arc welding in thia country. In the United States, however, alternating 
current haa been employed somewhat extensively, with, so we are given to imderstand, 
considerable success. A company known as The A.C. Cutting and Welding Company, 
Limited, the offices of which are at 25-27 Theobalds Road, Holbom, London, W.C. 1, 
has recently been formed to introduce into this country the machine and methods of 
on American firm. The machine, as the name of the company suggests, works with 
alternating current, and for alternating current, at any rate if used with its machine, 
the company claims numerous advantages. We may perhaps enumerate these claims 

Fio. 8).— The A.C. Electric Welding and Cutting Mwhine. 

before going on to describe the machine itself. First, there are low cost of operation 
and no maintenance chiirges. In this connection it is pointed out that the A.C. 
apparatus has no constantly moving parts, and that there is nothing, except accidental 
mechanical injury or abuse, to prevent its lasting indefinitely. The apparatus, it is 
stated, con8,umes approximately 0-15 kilowatt at a power factor of 65 per cent, when 
not in actual operation, and when working will deposit a pound of metal for from 1 to 2 
kUowatt'hours of energy. Then there is low cost of machine and wiring. On the first 
portion of this claim we make no comment, but as concerns the second portion it may 
be said that the distribution wiring is on the high-voltage side and that only small- 
sized wires are required as compared, say, with a direct-current system using a motor 



generator in which the distribution cables to the arc welders are necessarily of large 
cross section. The A.C. machine can be taken right up to the work, so that only quite 
short welding-circuit cables are required. 

Single PhdSe 

One mschlne on 
one ph^se ^tphtse 

Two machines on 2 ph^se 

One machine serosa 

outers of Zphdse 

te • 141 not 



One madiine 0n3 phsse ^efta, « 

Three machines on 3 phase deita 

One machine on 3 phase star 
*Thc Enoinccr" 

One machine connecte<f to neutral 
or 3 phase star-, voftaqe'SUotUne 

Three phase star' 2 machines to neutral 
'2 across regular power phases 

Three phase star^ 3 machines ha/anced 
en outers 

Three phsse'4mre sysiem-S machines teneuUml^S across outers-'^il 6a fenced 

Swain ^ 

Fia. 82. — Methods of Connecting up the A.C. Machinefl with Different Tjpes of Circuits. 

A third claim made for the A.C. mjLchine is that with it there is greater speed of 
deposition of the electrode metal than with other systems, and that no chipping or 
brushing of the work is necessary. The company maintains that alternating current 


provides a faster speed of welding than direct current when the electrode in the latter 
case is negative, because with the electrode negative approximately 60 per cent, of 
the heat is in the work, whereas with alternating current the heat is naturally divided 
equally between the work and the electrode. It is admitted that direct current with 
the electrode positive will provide a faster speed of welding, but it is urged that 
generally it is " too fast to be of any use . . . unless the arc is closed in by very heavy 
coating, so that the metal is enclosed in a viscous sleeve, or where the work is very 
rough, such as filling in with a large electrode on a casting where strength is not 
required." The following figures have been supplied to us of a test recently made with 
direct-current and alternating-current machines. In each case in which conditions of 
currents and voltage were strictly comparable it was found that alternating current 
was from 20 to 30 per cent, faster than direct current. Thus at 1 50 amperes and 25 volts 
4 lb. per hour of metal were deposited with alternating, as against 3 lb. per hour with 
direct current. At 175 amperes and 25 volts 5 lb. per hour were deposited with 
alternating current and 4 lb. per hour with direct current. As regards the contention 
that no chipping or brushing is necessary, it is certainly true that with the A.C. 
machine an arc can be struck from a rusty surface ; but we ;may say that the same 
thing can be done with direct current. 

A fourth claim for the A.C. system embodies " penetrtttion, smoothness and 
absence of electrolytic corrosion." It is asserted that an alternating-current arc has a 
penetrating power superior to that possessed by a direct-current arc, and that on that 
account there is set up a molecular diffusion extending from J in. to f in. below the 
surface of the work, so that its use is specially applicable to cast-iron welding. Again, 
the surface of the metal deposited by the A.C. arc is said to be smoother than that of 
metal deposited by direct-current arcs. Then with regard to electrolytic corrosion, it 
is pointed out that whereas with direct-current welding there is the choice of making 
the electrode either positive or negative, so that in either case there is a liability of the 
formation of a voltage couple with consequent electrolytic corrosion, especially in damp 
places, with alternating-current welding that liability is eliminated, as is also the chance 
of the deposition of dross in the weldj as may happen when the electrode is negative. 

Finally, the claim is made that the A.C. machine is more portable than any other 
and that the system is very flexible. It is pointed out that (a) the apparatus can be 
taken by two men exactly where it is required ; (6) the system is not tied down to a 
certain number of machines or certain positions ; (c) the system operates as efficiently 
with one machine running as with a few or a great many ; and (d) the machine, having 
no continually moving parts, will work in any position. 

Having set out the claims made for the system and apparatus, let us now con- 
sider the apparatus itself. Its outward appearance is shown in the engraving. 
Fig. 81. It consists of an oak box with louvred ends for ventilation, and with four 
arms like those of old-fashioned Sedan chairs for carrying purposes. As to the internal 
arrangements of the box, beyond the statement that they consist principally of a 
specially wound transformer the company prefers at the moment to give no definite 
particulars. All that we know for certain, therefore, is that the transformer has certain 
tappings, which will be again referred to later ; that there is a flux diverter, by means 
of which fine adjustments in the welding circuit are obtained ; and that the box also 
contains an electrically driven blower, the purpose of which is, by circulating a current 
of air through them, to keep the windings cool, and which absorbs some 60 watts. The 
complete apparatus is said to be weather-proof, so that it can be used out of doors as 
well as under cover, and skids are provided to facilitate movement of the box. The 
louvred openings are provided with fine-mesh screens. 


The primary leads are led out from the side of the box through insulated bushes. 
There are in all eight secondary plugs. They are arranged at one end of l^e box, and 
are motmted on a slate bed protected by an oak fronting. The plugs are split so as to 
make firm contact with the sockets at the ends of the welding leads. The eight welding 
tap plugs provide adjustment for various types of welding or for different sizes of 
electrode, any required adjustment within the limits of the machine being made by 
taking a socket off one plug and putting it on another, no other means of switching 
b^g necessary. The plugs may, in fact, be considered as being the means of bringing 
about what may be called the coarse adjustment of the welding circuit. The flux 
diverter, which is manipulated by the small hand wheel seen at the top of the box, and 
the movement of which can be observed through a circular glass window, is for fine 
adjustment after correct conditions have been roughly obtained by means of the plugs. 
It is explained, too, that the flux diverter also determines the depth of the penetrative 
effect of the arc — that is to say, whether"the metal will be sunk in as in sealing or added 
on as in padding. The flux diverter handle ia provided with an indicator dial to 
facilitate resetting when repetition work is being carried out. 

It will be observed that there are six primary leads. Only two are actually 
employed when the machines are at work, the others being carefully insulated, but the 

Fig. 83.— The A.C. Electrode Holder. 

Fig. 84.— The Holder Gripped to Eeleiwe Electrode. 

three sets are provided so as to enable the apparatus to be used on circuits the voltage 
of which is below the standard voltage for which the machine is intended. Moreover, 
should the supply voltage be higher than the voltage for the highest set of tappings on 
the primary, proper regulation can still be obtained by variation of the secondary plugs. 
The range of primary voltage with which any one apparatus can be successfully 
operated is therefore fairly wide. We return to this question later. Shifting the 
secondary leads on to plugs from left to right on either side of the lamp in the centre 
gives more heat in the weld. The function of the lamp, it may be explained, is to take 
up the " kick " of the arc when the circuit is opened, and any 1 10-volt lamp is suitable 



for the purpose. The^lamp also gives indication when the machine is ready for work. 
The A.C. machines are made in five sizes as follows : — 


Range In welding. 




Per cent. 

L.W. . . 

.. 40 to 110 




.. 75 to 175 




.. 100 to 350 



H.C.G. ... 

.. 300 to 650 



E.C.G. . . 

. . 500 to 1200 



The weights vary accordmg to the periodicities for which the transformers are wound. 
. The machine can be wound for any voltage, but it is not considered good practice to 
have a primary voltage in excess of 650. The machines are made as ordinarily wound 
for single phase for any voltage or frequency, but if occasion requires it they can be 
made for two or three-phase current, the power being taken from a two or three- 
phase circuit balanced on each phase, the welding circuit, of course, consisting of only 
two wires. The company informs us, however, that experience has taught it not to 
recommend these polyphase machines, as they cost much more and weigh much more, 
so that portability is lost with no attendant advantage of any kind. The diagram, 
Fig. 82, is interesting, as showing the methods of connecting the machine to various 
types of circuit. 

The standard machines are wound for one voltage only, such as 110, 220, 440, etc. ; 
but by reason of the means for connecting up the primary which are provided, and to 
which we have already referred, the 110-volt primary circuit can be 100, 105, or 110 ; 
the 220-volt primary circuit can be 200, 210, or 220 ; and the 440-volt primary circuit 
can be 400, 420, or 440. These figures correspond to 5 and 10 per cent, low to the 
nominated voltage. The secondary plugs provide a further range of 5 and 10 per cent, 

The power factor of the machine in welding is given as varying from 45 to 66 per 
cent., a fair average being 55 per cent., which means that the kilovolt-amperage 
required is approximately twice the kilowatts paid for. It is claimed that the load 
factor with the machine may be as high as 75 per cent, on the average, since there is no 
necessity to chip slag. Taking a power factor of 50 per cent, and a load factor of 
75 per cent., the cost of one machine per hour with energy at IJd. per kilowatt-hour 
is said to work out at less than 4d. In cutting, the power factor is so much higher 
and the load factor so much lower that the average demand in a given time is the same 
as when welding. The demand of the machine for the heaviest type of welding is 
said to be 5 kilowatts, and for cutting 10 kilowatts. 

In discussing the machine with the company's representatives emphasis was laid 
on the following points : '' A fundamental feature in all arc welding machines should 
be that a constant rate of heat should be automatically given out at the terminals. 
Constant rate of heat means constant energy, or, in other words, constant watts. I'his 
condition is brought about in our special transformer by a careful balance of the 
secondary and primary windings, so that as the voltage tends to increase, owing, say, 
to the arc lengthening or because of oil slag being in the path of the arc, the current 
tends to diminish and vice versa, so that energy is delivered at an approximately 
constant rate. 

**A fundamental essential for good welding is that a short arc be held. This 
condition is obtained in our machine by the swinging in and out of the voltage phase 
relations of three voltages in series in the secondary winding of the transformer. Such 
adjustment can be made that any short length arc can be held, and, when once set, 
the adjustment can be locked so that it is impossible for a longer arc to be drawn than 


the fixed maximum. This company considers it as beyond argument that a short 
arc is desirable in every case, for there is then less oxidation and nitrogenisation in the 
erater and less radiation than with a long arc, while there is more certainty of the 
electrode metal being deposited into the crater. 

" We are able to hold an A.C. arc with 31 volts open circuit, although in practice, 
because of line drop, poor contact, and various other reasons, we actually give more 
than that, taps on the machine providing different voltages, both essential — ^for keep- 
ing the arc going — and * kick ' voltages for different types of electrode and arc. 

*'In addition to the transformer supplying all these requirements automatically 
without continually moving parts, it further renders itself harm-proof when the 
secondary terminals are short-circuited, in that the voltage drops to zero with a current 
which can be adjusted to be less or greater than normal full-load current. As a matter 
of fact, we adjust it for slightly in excess of that current so that slightly greater heat 
is allowed for getting started. 

** The one characteristic of our machine which is not shared by self -regulating 
direct-current sets, which embody much the same characteristics of constant rate of 
energy in the welding circuit, is that each voltage can be changed — both * essential ' 
and * kick ' — ^without changing the current, whereas i^ith direct-current sets change in 
the voltage causes a corresponding change in the current." 

We have not, ourselves, seen any tests carried out on welds made by this apparatus, 
but we are informed that it will weld pieces which will bend flat on themselves — the 
bend being in the weld — and which in a twisting test may be twisted through two 
complete revolutions before tearing. 

The A.C. machine can be used with any type of electrode — bare wire, flux-coated, 
gaseous flux-coated, carbon, or graphite. The particular form of electrode holder or 
welding handle, as the company prefers to call it, recommended is illustrated in Figs. 
S3 and 84. It consists of a spring grip insulated holder, the metal portion of which is 
composed of a metal which is not affected by the heat of the arc. It enables all types 
of electrodes to be gripped at any desired angle without the necessity of keeping 
pressure on either handle. On a slight squeeze being given to the holder, as seen in 
Fig. 84, the stub end of a used electrode may be got rid of and a fresh electrode inserted. 
The current is led to the electrode through both halves of the handle equally. The 
flexible leads are brazed into the handle so that they become part and parcel of it. 


Abell, Mr. Westcott Stile, 97 
A.C. Cutting and Welding Co., Ltd., The, 117 
Electrode Holder, The, 120, 122 
Machines, Methods of Connecting up with 
different Types of Circuits, 118, 121 
Acetylene Welding, 2, 19 
" Added Thickness,*' 89 
Admiralty Authorities, The, 6, 13 
Aluminium, 15, 33 

for Flux for Electric WeJding, 33 
American Bureau of Steam Engineering, The, 

American Duplex " Spot " Welder, Large, 81 
Electrical Engineers, Listitute of, 71 
" Spot " Welders for Ship Work, 80 
Andrews, Mr. W. S. (American Genercd Electric 

Co.), 74 
A 1 Automatic Switch, 91 
" Butt " Welder, 93 
Manufacturing Co., Ltd., 89 et seq, 
"Spot *' Welders, 90 
Apparatus for Arc Welding, Machines and, 

49 et aeq. 
Ard between Two Electrodes:, Welding by means 

of, 3 
Arc, Crater of, Absorption of Oxygen by, 6 
Effect of, on Human Frame, 4 
Introduction of Gas into (Benaidos), 12 
Positive Crater of. Temperature of, 4 
Process, The Benardoe Carbon-, 8, 20 
Projected by Electro-magnet, 3 
The Plastic-, Welding System, 108 et seq. 
Use of for Making Holes (Benardos), 11 
Arc Welding, 1 et aeq., 8 et seq,, 33, 49, 65, 108, 1 17 

At a Steel Barrel Works, 19 e^ aeq. 
Company, Ltd., The British, 49 
Arc Welding Equipment, Self-contained (The 

British Westinghouse), 49 et seq. 
Arc Welding Equipment on Motor Lorry 

(British Westinghouse), 50 
Arc Welding Equipments, 50, 54, 65 
Arc Welding, Flux for, 5 

Arc Welding, Generator for (British Westing- 
house), 51 
Arc Welding, Glass Screens for Protection while, 

5, 20, 72 
Arc Welding, Helmet for, 4, 5 
Arc Welding, Loss in Steadying Resistemce 

during, 52 
Arc Welding, Machines and Apparatus for, 49 
et 9eq. 

Arc Welding, Motor Generator for (Crompton), 

Arc Welding, Poliffity of Electrodes for, 2 

Precautions to be tstken when, 4 
Screens for Use when, 5, 20, 71 

Arc Welding, Self-regulating Generators for, 
51, 53 et seq,, 62, 63, 64 

Arc Welding Set (Lancashire Dynamo and 
Motor Co., Ltd.), 56 

Arc Welding Set, Portable (J. H. Hohnes), 65, 66 

Arc Welding Set, Self-contained (Crompton), 72 

Arc Welding, SpecisJ Appliances Necessary 
for, 3 

Armour Plate, Formation of, by Electric 
Welding (Benardos), 12 

Asbestos, Blue, as Flux-forming coating for 
Electrodes, 33 

Attachments, "Butt" Welding (Al Manu- 
facturing Co.), 89, 94 

Attachments, " Seam " Welding (A 1 Manu- 
facturing Co.), 89, 95 


" Bad Metal," 6 

Bakelite, 81 

Balancer Set, ** Constant-energy," Electrical 

Connections of, 63 
Bafe Metal Electrodes, 3, 5, 71 
Barge, Electrically-welded, 5 
Barrel Co., Ltd., The Steel (Uxbridge), 19 et aeq., 

84 et seq. 
Barrel, Steel, Description of Making, 21 

Works, Arc Welding at a Steel, 19 
Benardos, Nicholas de, 1 et seq., 8, 9, 20 
and Olszewski Patent, 8 et seq. 
Carbon Arc Process, 8 et seq. 
Electrode Holders, 8 et seq. 
Machines for Arc Welding, 10 et seq. 
Bench Work, " Spot " Welder for (" Ponteleo "), 

BlockE, Radial Electrode-, "Butt" Welder 

with, 76, 77 
Blow-holes, The Filling up of, 36 
Blue Asbestos, as Flux-forming coating for 

Electrodes, 33 
Bonner, Mr. W. T. (Chester Shipbuilding Co., 

U.S.A.), 100 
Borax used as a Flux in Brazing, 29 
Bowling Iron Co., The, 3 
Brass, Welding, 15 
Brazing Collars on Tubes, Meu^hine for, 28 



Bridge Welding (Pontelec), 27 

British Electric Plant Ck)., Ltd., 49 

British Insulated & Helsby Cables Ltd. (B.I.W.), 

39 ei seq. 
B.I.W. R^istance Welders, 39 et seq. 
" Seam " Welders, 45, 46 
" Spot " Welders, 47, 48 
Wire Welders, 39 
British Westinghouse Electric & Manufacturing 

Co., Ltd., 49 
Bruce Peebles & Co., Ltd., 49 
Bubbles, Gas, in Electric Welds, 6 
Burd, Rear Admiral, U.S.N., 111 
Bureau of Steam Engineering, The American, 

" Butt " Welder (Electric Welding Co., Ltd.), 79 
'* Butt " Welder for Miscellaneous Repair Work 

(Electric Welding Co., Ltd.), 77 
" Butt " Welder for Hoops, Rings, etc. (Electric 

Welding Co., Ltd.), 79 
" Butt " Welder for Hoops, Rods, etc. (Electric 

Welding Co., Ltd.), 76, 76 
" Butt " Welder for Tyres (Pontelec), 31, 
" Butt " Welders (Al Manufacturing Co., Ltd.), 

89 elaeq. 
" Butt " Welders (B.I.W.), 39 et seq. 
*• Butt " Welders (Electric Welding Co., Ltd.), 

75 et aeq, 
•* Butt " Welding, 2, 17. 31. 39 et aeq. 
"Butt" Welding Attachments (A 1 Manu- 
facturing Co., Ltd.), 89, 94 
" Butt " Welding of T^res, Diagram Showing 

Principle of, 17 

Caldwell, Capt. (afterwards Major) James, 
4, 6, 99 et aeq. 

Caps for Electrodes of Large Welders, 83 

Candy, Mr. A. M., 70, 71 

Ceurbon, Alteration of Composition of Welded 
Metal by, 2 

Carbon Arc Process, The Benardos, 8 
Welding, 6, 20 

Carbon Arc Welding, First Employed Com- 
mercially, 2, 3 

Carbon Electrode Holder, 20 
Electrodes, 2, 3 

Carbon Electrodes Cored, 3 

Cast Iron, Welding, 36 

Chain-link-forming Machine, B.I.W., 41 

Oiain-Hnk Welder, B.I.W., 43, 44 

Chains, Machines for Making, B.I.W., 41 6/ aeq. 

Clapper, Mr. J. B. (Standard Parts Co., Cleve- 
land, Ohio), 14 

Chester Shipbuilding Co., U.S.A., 100 

Clutch, The Davies & Soewnes (Daysohms) 
Electro-magnetic, b^ et aeq. 

Coated Electrodes, 5 et aeq., 33 ei aeq. 

Coating, Slag-forming, for Electrodes, 6 

Coatings for Electrodes, aU not Slag-forming, 6 

Collars, Machine for Brazing on Tubes, 28 

Connections, Electric, for A.C. Welding 
Machines with different Types of Circuits, 1 18 

Connections, Electric, of British Westinghouse 
Generator for Arc Welding, 51 

Connections, Electric, of " Constant-energy " 

Balancer Set, 63 
Connections, Electric, of Davies & Soames 

(Dayaphms) Differential Electro-magnetic 

autch, 61, 62 
Connections, Electric, of Lancashire Dyncimo & 

Motor Co.'s Arc-welding Set, 57 
Connections, Electric, of Large American 

Duplex *' Spot '* Welder, 82 
'* Constant-energy " Arc- welding Set, The, 62, 

Continuous or " Point " Welding Machine 

(Benardos), 11 
** Continuous-union " Welding (Benardos), 9 
Converter, Rotary, 96 
Converter, Rotary Motor, 64 
Cored Carbon Electrodes, 3 
Coming Glass Co. of New York, 74 
Crater of Positive Arc, Temperature of, 4 
Crompton & Co., Ltd., 69 et aeq, 
Crompton Self-contained Arc Welding Set, 72. 
Crookes', The late Sir WilUam, Glass, 21, 72 
Current Limits admissible with different sizes 

of Quafei-arc Electrodes, 36 
Current Variation in Arc Welding, 4, 22 
Currents, Primary and Secondary, in Large 

American Welders, 80 
Cutting & Welding, The A.C. Co., 117 
Cutting & Welding, The A.C. Co. Machine, 117 
Cutting of Metals, The, 11, 37, 112 


Damage to German Steamships, 108 

Davies & Soame^ (Daysohms) Electro-magnetic 

Differential Clutch, 58 et aeq, 
Davies, Mr. Walter L., 68 
Defects in Electric Welds, 6 
de Meritens, 8 
Device for Continuous or " Point " Welding 

(Benardos), 11 
Device for " Point " Welding (Benardos), 10 
Dickinson, Mr. (The Bowling Iron Co.), 3 
Differential Clutch, The Davies & Soames 

(Daysohms) Electro-magnetic, 58 et aeq* 
Disc Welding (" Pontelec "), 27 
Disunion of Metals (Benardos), 11 
Drills, Machine for Welding (B.I.W. ), 41 , 
Drums, Oil-, Manufacture of, by Electric 

Welding, 84 
Duplex "Spot^* Welder, Lckrge American, 81 
Duplex " Spot " Welder, Large American, 

Electriccd Arrangements of, 82 
Dynamo & Motor. Co., Ltd.» The Lancctshire, 

49, 56 e^ aeq. 


Earthing Wire for Welding Machine, 91 
Effect of the Arc on the Human Frame, 4 
Elastic-limit of Electric Welds, 98 
Electric Arc Welding, Benardos*, Machines for, 

10, 11 
Electric Forcje, Benardos', 10 

Plant Co., Ltd., The British, 49 

Welding Co , Ltd., The, 49, 75 et aeq. 

Welding Co., Ltd., The Premier, 49 

et aeq. 



Electric Welds, Defects in, 6 

Examples of (Benardos), 9 
Electrical Arrangements of Large American 

Duplex " Spot " Welder, 82 
Electrical Engineers, American Institute of, 71 
Electrically-welded Barge, 6 
Electro-magnetic Clutch, Da vies & Soames 

(Daysohms), 58 et acq. 
Electrode Holder, The A.C. Cutting & Welding 

Co., Ltd., 120 
Electrode Holder, The Benardos, 9 
Electrode Holder, The Equipment & Engineer- 
ing Co., 70, 73 
Electrode Holder, The T. T. Heaton, 20 

Holdeis, American, 70, 71 
Electrodes, Bare Metal, 3, 5, 71 
Benardos, 2, 3 
Carbon, 2 

Coatings for, not all Slag-forming, 6 
Cored Carbon (Bencurdos), 3 
Flux Coated, 5, 33 
Graphite, 20, 71 
Heat of Positive and Negative, 2 
Kiellberg, 6 
Electrode?, Mild Steel, suggested by Mr. T. T. 

Heaton, 3 
Electrodes, Polarity of in Arc Welding, 2 

Quasi-arc, 35 
Electrodes, Quasi-arc, Sizes for Sheet and Plate 

Welding, 37 
Electrodes, Slag-forming Coatings for, 6, 33 

Slavianoff, 3 
Emergency Fleet Corporation, The, U.S.A., 100 
" Energy, Constant-", Arc Welding Set, 62, 63 
Equipment and Engineering Co., The, 58, 64, 70 
Eye Protection during Arc Welding, 71 et seq. 

Filling up Blow-holes, 36 

Fleet Corporation, The Emergency, U.S.A., 

Flux for Arc Welding, 5, 6, 22, 29 

" Flux, Gaseous," I^ocess, 6 

Forge, Electric, Benardos', 10 

Formation of Armour Plate, Benardos' Sugges- 
tion for, 12 

Frame, Human, Effect of Arc on, 4 

SpectcMsie-, Soldering Machine, 29 

Franklin Institute of Philadelphia, 14 

Friedrich der Oroaae, Damage to German Liner, 
110, 111 


Ga9 Bubbles in Electric Welds, 6 

Gas, Litroduction into the Arc, proposed by 

Benardos, 12 
General Electric Co. of America, 74 
Generating Set for Arc Welding r. Portable, J. H. 

Holmes, 66, 66 
Generators for Arc Welding, Self -regulating, 51, 

53 et seq., 62, 70 
German Steamships in America, Damage to by 

Crews, 108 
Glass, Coloured, for Plx>tecting the Eyes, 12, 20, 


Glass, Crookes', 21, 72 

Glass, Gold-plated, for Eye-protection during 

Arc Welding, 74 
Glass, " Safety," for Arc Welding, 72, 73, 74 

Screens for Protecting the Eyes, 11, 20, 

72, 73 
Grand Trunk Railway, Repairs on, 115 
Graphite Electrodes, 20, 71 


Hand-operated Welder (Electric Welding Co.), 
77, 79 

Heaton, Mr. T. T., 19, 20, 21, 84 

Helmet for Arc Welding, 4, 5 

Holder, Electrode, American, 70, 71 

The A.C., 120 
The Benardos, 9 

Holder, Electrode, The Equipment and Engin- 
eering Co., 73 

Holder, Electrode, The T. T. Heaton, 20 

Holmes, J. H., & Co., 49, 65 et seq. 

Hoops, Rings, etc., " Butt " Welder for 
(Electric Welding Co.), 78, 79 

Human Frame, Effect of Arc on, 4 

Huron, S.S. (ex-German Friedrich der Grosae), 
Damage to. Ill 

Hydraulic Pressure for Consolidating Welds, 75 

Iron, Welding Cast, 36 

Jessop, Capt. E. P. (U.S.N.), 111 
Jevons, Mr. A., 25 

Jews' Harp Welder (Elihu Thomson), 14 
Joint, Welded-riveted, 35 
Joints, Welded and Riveted, Comparative 
Strength of, 101 


Kaiser Wilhelm II, German Liner, Damage to, 

Kjellberg, Coated Electrode, 6 

" Gaseous-flux " Process, 6 
Koester, Capt. O. W. (American Bureau of 

Steam Engineering), 111 

Lancashire Dvnamo and Motor Co., Ltd., 49, 

56, 57 
Large ** Spot " Welders (American General 

Electric Co.), 78, 80 et seq. 
Large "Spot " Welding Machine, A (Pontelec), 

Line Welding, 2 

Link Welder, Chain (B.I.W.), 43, 44 
Lloyd & Lloyd, Ltd., 3 
Lloyd's Register, 1, 97 
Lloyd's Register, Tentative Regulations for 

Electric Welding, 97 
Locomotive Frame, Cracked, Repair of, 116 



Lorry, Motor, Arc Welding Equipments 
(Til ling-Stevens), 65 et seq. 

Lorry, Motor, British Westinghouse Arc Weld- 
ing Equipment on, 49, 60 

Loss in Steadjong Resistances During Arc 
Welding, 52 


Machine, A Large British " Spot " Welding, 103 
Machine, Arc Welding (Benardos), 3, 4 
Machine, Chain-link-forming (B.I.W.)f 41 
Machine for Brazing Collars on Tubes (Pontelec), 

Machine for Soldering Spectacle-frames (Pon- 
telec), 20 
Machine for Welding Drills (B.I.W.), 41 
Manufacture of Chains, Machines for (B.I.W.)> 

41 et aeq. 
Marquand, Mr. H. S., 51 
Mather & Piatt, Ltd., 49 
Mavor & Coulson, Ltd., 49 
Metal, " Non-arcing," 92 
Metals, Cutting, by the Quasi-arc Process, 37 
Metals, Separation or Disunion of, Benardos, 1 1 
Mild Steel Electrodes suggested by Mr. T. T. 

Heaton, 3 
Miscellaneous Repair Work, ** Butt " Welder 

for (Electric Welding Co.), 77 
Modulus of Elasticity of Electric Welds, 98 
Motor Lorry, Arc Welding Equipments (Tilling- 

Stevens), 65 ei acq. 
Motor Lorry, British Westinghouse Arc Welding 

Equipment on, 49, 50 
Motor Generators for Arc Welding, 4, 49 e< seq., 

70, 73 

'Nassovia, German Liner, 108 
Navy, U.S., Secretary of, 108, 111 
Noble, Mr. P. O. (American General Electric 
Co.), 62 


Oil Drums, Manufacture of, by Resistance 

Welding, 84 
Oil Drums, Resistance Welders for the Memu- 

facture of, 87 
Olszewski, Benardos and. Patent of, 8, 20 

Stanilaus, 8 
Operators, Training of. Time taken in, 21, 89 
Overhecki Welding, 36 

Patent, Benardos and Olszewski, 8 

Peebles, Bruce & Co., Ltd., 49 

Peters, G. D. & Co., Ltd., Ill 

Phoenix Dynamo Mcmufacturing Co., Ltd., 25, 

Plcustic-eirc Welding and Cutting Panel, 113 

System, The, 108 et seq. 
Plate and Sheet Welding, Sizes of Quasi-arc 

Electrodes required for, 37 
Plate Armour, Suggested Formation of, 

Benardos, 12 

Pocahontas, Repaired and Re-named German 

Liner, 111 
" Point-union," 9 
Point, Multiple, Welding, 27 
Polarity of Electrodes in Arc Welding, 2 
Pontelec Machine for Brazing Collars on Tubes, 

Pontelec Methods and Machines, 25 
Resistance Welder, 26 
" Spot " Welder, Large, 103 
Typical, " Seam " Welder, 31, 32 
Welding Patents, Ltd., 25 et seq,, 103,. 
Pontelec Welding Processes, 27 et eeq. 
Portable Welding Set (Crompton), 72 

{J. H. Holmes), 65, 66 
Portable Welding Set (Premier Electric Weld- 
ing Co., 53 et 8eq, 
Precautions to be taken when Arc Welding, 4 
Premier Electric Welding Co., 49, 53 et seq. 
Premier Welding Set in Cabin, 54 
Pressure, Hydraulic, for Consolidating Welds, 75 
Prinz Joachim, German Steamship, Damage to, 

Prinzess Irene, German Steamship, Damage to, 

Process, The Benardos Carbon Arc, 8 

The Quasi-arc, 33 et seq. 
Processes, Pontelec Welding, 27 

Quasi- ABC Co., Ltd., 31 ef seq. 

Electrodes, 35 
Quasi-arc Electrodes, Current Limits Admis- 
sible with Different Sizes of, 36 
Quasi-arc Electrodes, Sizes Required for Sheet 

and Plate Welding, 37 
Quasi-arc Process, The, 33 et seq. 

Cutting Metals by, 37 


Radial Blocks, " Butt " Welder with, 76, 77 
Railway, Grand Trunk, Repairs on, 115 
Rawson, Woodhouse &, 25 
Reactance in Arc Welding Circuit, 62 
Rees-Ro turbo Manufacturing Co., Ltd., The, 49 
Regulating, Self-, Generators for Arc Welding, 

51, 53 et seq., 62, 70 
Reinforcing Worn or Weak Work, 36 
Repair Work, Miscellaneous, " Butt " Welder 

for (Electric Welding Co.), 77 
Repair Work, Railway, 115 
Resistance Frame for Arc Welding (J. H. 

Holmes), 65, 67 
Resistance, Steadying, for Arc Welding, 3, 33, 

34, 52, 65, 67 
Resistance, Steadying, Loss in, 52 
Resistance Welder, Caps for Electrodes of Large, 

Resistance Welders (Al), 89 et seq. 

(B.i.W.), 39 et seq. 
Resistance Welders, Electric Welding Co., Ltd., 

15 et seq. 
Resistance Welders for Oil-drum-making, 87 



Hesistcuice Welders, Leirge American " Spot," 
78, SOetseq. 

Resistance Welders (Pontelec), 26 et seq., 103 

Resistance Welding, 1, 14, 15, 25, 38, 39, 75, 89 

Resistance Welding, Discovery of, by EHhu 
Thomson, 14 

Resistance Welding, Manufacture of Oil-drums 
by, 84 

Resistance Welding, Voltage for, 15 

Ridge Welding (Pontelec), 28 

Riveted and Welded Joints, Comparative 
Strength of, 101 

Rotary Converter (A 1 Manufacturing Co.), ®6 

Rotsury Treucisformer (Equipment and Engineer- 
ing Co.), 64 

Roturbo Msuiufacturing Co., Ltd., The Rees, 49 





SOBEEXS for Arc Welding, 5, 20, 71 

Seam " Welder, Longitudinal (A 1), 89 
Seam " Welder, Tyjiical Pontelec, 31, 32 
Seam Welding, 3, 9, 17, 31, 46 et seq. 

Welding Attachments (Al), 89, 95 

Seam " Welding, Diagram Showing Principle 

of, 17 
" Seam " Welders, Examples of B.I.W., 45, 46 
Self-contained Arc- welding Set (Crompton), 73 
Self -regulating Generators for Arc Welding, 61, 

53 et seq,, 62, 70 
Separation or Disunion of Metals (Benardos), 11 
Sheet and Plate Welding, Sizes of Quasi-arc 

Electrodes required for, 37 
Ship Work, American ** Spot " Welders for, 

Skill, Operator's, Percentage Factor in Success- 
ful Welding, 21 
Slag-forming Coating for Electrodes, 6, 33 
Slavianoff, 3 
SoBjnes, Mr. Alfred, 58 
Spectacle-frames, Machine for Soldering 

(Pontelec), 29 
" Spot " Welder (A 1 Manufacturing Co.), 90 

for bench Work (Pontelec), 30 
** Spot " Welder, Large American, for Ship 

Work, 80 
** Spot " Welder, Large American Duplex, 81 

Typical Pontelec, 30 
Welders (B.I.W. ), 47, 48 
Welding, 2, 9, 16, 26, 27, 30, 31, 47, 
et seq.y 78, 80 et eeq,, 90 et seq., 103 et seq. 
*' Spot " Welding, Diagram Showing Principle 

of, 16 
" Spot '* Welding Machine, A Large British 

(Pontelec), 103 et aeq. 
Standard Parts Co., Cleveland, Ohio, 14 
Steadying Resistance for Arc Welding, 3, 33, 34, 

52, 65, 67 
Steel Barrel Co., Ltd., Txbridge, 3, 19, 84 

Works, Arc Welding at, 19 
Steel Barrels, Description of making of by 

Carbon Arc Welding, 21 
Stewarts & Lloyd, Ltd., 3 
Strength of Electric Welds, 97 
Strips, '* Butt " Welder for (Electric Welding 

Co.), 79 

Strohmenger, Mr. Arthur Percy (Quetsi^arc), 33- 

Switch, Automatic (" A 1 "), 91 

Sjrstem, The Benardos Arc Welding, 8 et aeq. 
The Plastic-arc Welding, 108 
of Arc Welding, The A.C., 117 
of Arc Welding, The Benardos, 8 et aeq. 
Thomson Welding, Diagram of, 16 

Temperature of Arc, 4 
" Thickness, Added," 89 
Thomson Electric Welding Co., The, 15 
Ehhu, 1, 14 

-Houston Electric Co., 14 
Welding System, Dictgram of, 16 
Tilling-Stevens, Ltd., 65 et aeq. 
Traction Work, Rotary Transformer for Weld- 
ing Work in, 64 
Trckining of Welding Operators, Time taken in,. 

21, 89 
Transformer, Arrangement of, for Large 

American " Spot " Welder, 81, 82 
Transformer, Arrangement of, for Large 

Pontelec " Spot " Welder, 106, 107 
Transformer, Rotary, for Trewtion Arc Welding 

Work, 64 
Transformer, Static (Electric Welding Co.), 75 
Transformers, Regulating, 80 
Transformers, Static, 3, 14 e^ aeq.^ 16, 75, 80, 8U 

82, 106 
Tubes, " Butt " Welder for (Electric Welding 

Co.), 75, 76 
Tubes, Machine for Brazing Collars on 

(Pontelec), 28 
Typical Pontelec " Seam " Welder, 31, 32 

" Spot " Welder, 30 
Tyres, ** Butt " Welding of, Dictgram showing 

principle of, 17 

Ultimate Strength and Elongation of Welded 
Material, 98 

Union, Continuous-," Welding (Benardos), 9 
Union, Point-," Welding (Benardos), 9 
United States Shipping Board, 100 
Uxbridge, Steel Barrel Co.'s Works at, 19, 84 

Variation of Current in Arc Welding, 4 
Vaterland, German Liner, Damage to, 108 
Voltage for Arc Welding, 2, 24, 33, 61, 111, U5^ 

Voltage for Plastic-arc Welding, 111, 115 
of A.C. Co.'s Machines, 121 


Waahington, Oeorge, German Liner, Damage to, 

Weld, Absorption of Oxygen by, 5 
Welded Barge, Electrically, 5 

and Riveted Joints, Comparison of, 101 
Welded Material, Ultimate Strength euid 

Elongation of, 98 



Welder, " Bait " (B.I.W.), 40, 43 

Welder, " Butt/' for Tubes (Electric Welding 

Co.), 75, 76 
Welder, '' Butt," with Radial Welding Blocks 

(Electric Welding Co.), 76, 77 
Welder, Chain Link (B.I.W.), 43, 44 

Hand-operated (Electric Welding Co.). 
Welder, Jews' Harp (Elihu Thomson), 14 

Large American Duplex ** Spot," 81 
Resistance, T3rpical Pontelec, 26 
Welder, " Spot " for Bench Work (Pontelec), 

Welders, "Butt" (A 1 Manufacturing Co., 

Ltd.), 89 
WeldeFB for Longitudinal Seams, 31, 32, 87 
Welders, " Spot," for Ship Work, American, 80 

for Steel barrel Mcmufetcture, 
Resistance (Electric Welding Co.), 
75 et seq. 
Welders, " Seam " (B.I.W.), Examples of, 45, 46 
" Spot " (A 1), 89, 90 

(B.I.W.), 47, 48 
Wire (B.I.W.), 39 
Welding, Acetylene, 2, 19 
Welding, Arc, 1 et seq., 14 el seq,, 19 e< seq., 33, 

49, 65, 108 et seq„ 117 
Welding, Arc, at a Steel Barrel Works, 19 
Welding, Arc, Electrode Holders for, 9, 20, 70, 

71, 73, 120 
Welding, Arc, Helmet for, 4, 5 
Welding, Arc, Machines and Apparatus for, 

49 et seq. 
Welding, Arc, Precautions to be tedcen when, 4 
Welding, Arc, Set, The " Constant -Energy," 

62, 63 
Welding, Arc, Set, Petrol-driven, Self-con- 
tained (Crompton), 70 
Welding, Arc, Set, Petrol-driven, Self-con- 
tained (J. H. Holmes), 65, 66 
Welding, Arc, Set (Lancashire Dynamo and 

Motor Co.), 56, 57 
Welding, Arc, Steadying Resistance for, 3, 33, 

34, 52, 65, 67 
Welding, Bridge, 27 

Welding by Arc Struck between Two Elec- 
trodes, 3 
Welding, "Butt," 2, 17, 31, 39 et seq., 76 

et seq., 89 et seq. 
Welding, "Butt," of Tyres, 17, 31 
Cast Iron, 36 

Welding, " Continuous," or " Point," Machine 
for (Benardos), 11 
- Welding, " Continuous-union " (Benardos), 9 
Disc (Pontelec), 27 
Drills, Machine for (B.LW.), 41 
Welding Hoops, Machine for (Electric Welding 

Co.), 78, 79 
Wekling, Line, 2 

Machine, a Large British ** Spot," 103 
Machines, Examples of Benardos, 10, 

Midtiple-point (Pontelec), 27 
Overhead, 36 

Polarity of Electrodes in Arc, 2 
" Point " Machine for (Benardos), 10 
«' Point-union " (Benardos), 9 
Portable Arc Plant (Plremier), 54, 55 
Processes of (Pontelec), 27 
Premier Electric Co., The, 49 
Welding, Resistonce, 1, 14 et seq., 25, 39, 75 

et seq., 84 et seq., S9 et seq 
Welding, Resistance, Discovery of, by Elihu 

Thomson, 14 
WeWing, " Ridge " (Pontelec), 28 

" Seam," 2, 9. 17, 32. U et seq. 
Welding, " Seam," Diagram showing Principle 

of, 16 
Welding, Sheet and Plate, Sizes of Quasi-arc 

Electrodes required for, 27 
Welding, " Spot," 2, 9, 16, 26, 27, 30, 31, 47 

ei seq., 78 et seq., 84 et seq., 89 et seq., 103 
Welding, " Spot," Diagram showing principle 

of, 16 
Welding System, The Thomson, Diagram of, 16 
The Electric Co., Ltd., 49 
Thomson, System of, 14, 25 
Welds, Electric, Defects in, 6 

The Strength of , 97 et seq., 115 
Gas Bubbles in, 6 
Westinghouse Co., The British, Arc Welding 

Equipment for Motor Lorry, 49, 50 
Weyman, Mr. J. E., 65 
Wilson System of Welding, List of Vessels 

Repaired by, 112 
Wilson System of Welding (Plastic-arc), 111 
Wilson Welder and Metals Co., 108 
Wire, Earthing, 91 

Welders (B.I.W.), 39 
Woodhouse & Rawson, 25 

Zerener of Leipzig, 3 





TIlia book may bt ktpt 


from lact dst« sUmiMd bdow. A fine of TWO CENTS 
will bo obAiiMt for Meb day tbo book U koirt ot« 

'i J-'!??^